CN100538498C - Object determination device and camera head - Google Patents

Abstract

Translated fromChinese

仅对根据预定的任意(规定)焦点信息所摄像的图像执行脸检测。在该脸检测时，根据因人脸的凹凸或器官而产生的多个特征区域内的统计量的相对值来检测人脸。因此，即使在供脸检测用的图像中脸被拍得模糊，也能取得某区域内的统计量，因而可检测该脸。因此，在摄像装置中在预备摄像阶段不进行焦点控制，可快速地检测出成为对焦对象的主要被摄物体。

Face detection is performed only on images captured based on predetermined arbitrary (prescribed) focus information. In this face detection, a human face is detected based on relative values of statistics in a plurality of feature regions due to unevenness or organs of the human face. Therefore, even if a face is blurred in an image for face detection, the statistics in a certain area can be obtained, so that the face can be detected. Therefore, in the imaging device, focus control is not performed in the preliminary imaging stage, and the main subject to be focused can be quickly detected.

Description

Translated fromChinese

对象决定装置和摄像装置Object determination device and camera device

技术领域technical field

本发明涉及可有效应用于摄像静态图像和动态图像等的装置(数字相机(digital still camera)和数字摄像机(digital video camera)等)的技术。The present invention relates to technologies that can be effectively applied to devices (digital still cameras, digital video cameras, etc.) that capture still images and moving images.

背景技术Background technique

在摄像装置中配备有对特定的被摄物体进行对焦的自动聚焦、和根据特定的被摄物体执行曝光控制的自动曝光控制等技术。以往，对在帧内预先规定的特定位置(例如帧内的中央点)存在的被摄物体实施自动聚焦或自动曝光控制。然而，根据用户的成帧(framing)喜好，用户想关注的被摄物体，即想作为自动聚焦或自动曝光控制的对象的被摄物体并不限于位于帧的中央。在这种情况下，用户操作摄像装置以使想关注的被摄物体移动到帧内的特定位置，在该条件下进行了自动聚焦或自动曝光控制后，必须实施与喜好对应的成帧，进行这种处理是极其麻烦的作业。The imaging device is equipped with techniques such as automatic focusing for focusing on a specific subject, and automatic exposure control for performing exposure control according to a specific subject. Conventionally, autofocus or autoexposure control is performed on a subject existing at a predetermined specific position within a frame (for example, a central point within a frame). However, according to the user's framing preference, the subject that the user wants to focus on, that is, the subject that the user wants to be the subject of autofocus or autoexposure control is not limited to being located at the center of the frame. In this case, the user operates the imaging device so that the subject he wants to pay attention to moves to a specific position within the frame, and after performing autofocus or autoexposure control under this condition, it is necessary to implement framing according to his preference, and perform Such processing is extremely troublesome work.

作为解决该问题的技术，有为了确定要进行对焦的对象或曝光控制对象而预先实施摄像(以下把这种摄像称为“预备摄像”)的技术。在进行预备摄像的技术中，使用在预备摄像中所摄像的图像来确定对焦或曝光控制的对象。As a technique for solving this problem, there is a technique of performing imaging in advance in order to specify a subject to focus on or an exposure control target (hereinafter, this imaging is referred to as “preliminary imaging”). In the technique of performing preliminary imaging, a target for focus or exposure control is specified using an image captured in the preliminary imaging.

作为进行预备摄像的技术的具体例，有与视场(图像内)的人物的位置/范围无关地对图像内的人物的脸进行对焦的技术(参照专利文献1)。同样，有与图像内的人物的位置/范围无关地对图像内的人物进行合适曝光的技术(参照专利文献2)。As a specific example of the technique of performing preliminary imaging, there is a technique of focusing on the face of a person in the image regardless of the position/range of the person in the field of view (in the image) (see Patent Document 1). Similarly, there is a technique for appropriately exposing a person in an image regardless of the position and range of the person in the image (see Patent Document 2).

在上述现有技术中，当在画面内检测出多个脸时，把距中央部近的脸的至少一部分作为测距区域来进行自动对焦控制。并且，作为同样在画面内检测出多个脸的情况下的技术，也具有自动地使画面内的全体人员进入景深的技术(参照专利文献3)。In the prior art described above, when a plurality of faces are detected in the screen, at least a part of the face closer to the central portion is used as a range-finding area to perform autofocus control. Also, as a technique for detecting a plurality of faces in a screen, there is a technique for automatically bringing all people in the screen into the depth of field (see Patent Document 3).

专利文献1：特开2003-107335号公报Patent Document 1: JP-A-2003-107335

专利文献2：特开2003-107555号公报Patent Document 2: JP-A-2003-107555

专利文献3：特开2003-092699号公报Patent Document 3: JP-A-2003-092699

发明内容Contents of the invention

然而，在现有技术(例如专利文献1～3所述的技术)中存在以下问题。即，存在的问题是，如果在通过预备摄像获得的图像中最初焦点相对于主要被摄物体大幅偏离，则存在不能检测主要被摄物体即脸的可能性。However, there are the following problems in the prior art (for example, the techniques described inPatent Documents 1 to 3). That is, there is a problem that if the focus of the image obtained by the preliminary imaging is largely deviated from the main subject at first, there is a possibility that the face which is the main subject cannot be detected.

作为解决该问题的方法，有通过在预备摄像阶段进行焦点控制，来摄像可检测到主要被摄物体的图像的方法。具体地说，例如有从通过预备摄像获得的多个图像中检测主要被摄物体的方法。在该方法中，在预备摄像时摄像多个已进行了对焦的图像。然后，根据该多个图像实施主要被摄物体的脸检测。因此，即使在某图像中由于主要被摄物体的脸模糊而不能检测的情况下，也能在其它图像中检测出主要被摄物体的脸。As a method of solving this problem, there is a method of capturing an image in which a main subject can be detected by performing focus control in a preliminary imaging stage. Specifically, for example, there is a method of detecting a main subject from a plurality of images obtained by preliminary imaging. In this method, a plurality of focused images are captured during preliminary imaging. Then, face detection of the main subject is performed based on the plurality of images. Therefore, even in a case where the face of the main subject cannot be detected because the face of the main subject is blurred in a certain image, the face of the main subject can be detected in the other image.

然而，在这种方法，必须对多个图像执行脸检测。并且，为了在摄像装置中摄像多个已对焦了的图像，有必要进行焦点调节。因此，与现有的方法相比，存在对焦处理需要时间的问题。However, in this approach, face detection must be performed on multiple images. In addition, in order to capture a plurality of focused images with the imaging device, it is necessary to perform focus adjustment. Therefore, compared with the conventional method, there is a problem that focusing processing takes time.

在本发明中，第一目的是解决该问题，在摄像装置中在预备摄像阶段不进行焦点控制，快速检测成为对焦或曝光控制的对象的主要被摄物体。In the present invention, the first object is to solve this problem by rapidly detecting a main subject to be subject to focus or exposure control without performing focus control in a pre-imaging stage in an imaging device.

另外，在通过进行上述预备摄像来决定对焦或曝光控制的对象的情况下，存在以下问题。在预备摄像的图像内所检测出的人脸的数目不一定是一个。在所检测出的人脸的数目是一个的情况下，只要无条件地对该脸进行对焦或曝光控制即可。然而，在预备摄像的图像内检测出多个人物的脸的情况下，应对哪个脸进行对焦或曝光控制，迄今为止尚未作过充分考虑。例如，具有根据在帧中央所拍摄的人脸是用户期望作为被摄物体的脸的方针，对距帧中央最近的人脸进行对焦或曝光控制的技术。然而，这种技术是以用户最期望作为被摄物体的脸在帧内为一个这样的前提为基础的，如果在帧内存在多个被同等期望作为被摄物体的脸，则不能可靠地应对。In addition, in the case where the subject of focus or exposure control is determined by performing the above-mentioned preliminary imaging, there are the following problems. The number of human faces detected in the pre-captured image does not necessarily have to be one. When the number of detected human faces is one, it is only necessary to unconditionally perform focus or exposure control on that face. However, when a plurality of human faces are detected in a pre-captured image, which face should be focused or exposed has not been fully considered. For example, there is a technique of performing focus or exposure control on the face closest to the center of the frame based on the guideline that the face captured at the center of the frame is the face desired by the user as a subject. However, this technique is based on the premise that there is only one face in the frame that the user most expects as the subject, and cannot reliably cope with multiple faces that are equally expected as the subject in the frame. .

在本发明中，第二目的是解决该问题，提供在预备摄像中检测出多个人物的脸的情况下，判断应对哪个脸进行对焦或曝光控制的装置。The second object of the present invention is to solve this problem, and to provide an apparatus for determining which face should be focused or exposed when a plurality of human faces are detected during preliminary imaging.

〔第一方式〕[First way]

在以下说明中，肤色意味着所有人的肤色，不限于特定人种的肤色。In the following description, skin color means the skin color of all people, not limited to the skin color of a specific race.

为了达到上述第一目的，本发明采用以下的结构。本发明的第一方式是一种对象决定装置，包括脸检测单元和决定单元。In order to achieve the above-mentioned first object, the present invention employs the following structures. A first aspect of the present invention is an object determination device including a face detection unit and a determination unit.

脸检测单元把根据任意的焦点信息所摄像的图像作为处理对象的图像，从前述处理对象的图像中检测人脸。任意的焦点信息可以是规定的焦点信息。这里，规定的焦点信息是预先定为一个的焦点信息，而不是动态变化的焦点信息。因此，在根据规定的焦点信息来摄像图像的装置，即摄像被输入到脸检测单元中的图像的摄像装置中，期望的是预先实施基于该规定的焦点信息的焦点控制。例如，期望的是，根据该焦点信息预先实施聚焦镜头等的控制。The face detection means takes an image captured based on arbitrary focus information as an image to be processed, and detects a human face from the image to be processed. Arbitrary focus information may be prescribed focus information. Here, the predetermined focus information is predetermined focus information rather than dynamically changing focus information. Therefore, in an imaging device that captures an image based on predetermined focus information, that is, an imaging device that captures an image input to a face detection unit, it is desirable to perform focus control based on the predetermined focus information in advance. For example, it is desirable to perform control of the focus lens and the like in advance based on the focus information.

通过这样预先控制摄像装置，就没有必要在摄像被输入到脸检测单元中的图像时重新实施焦点控制。因此，摄像被输入到脸检测单元中的图像所需要的时间缩短。By controlling the imaging device in advance in this way, it is not necessary to re-implement focus control when imaging an image input into the face detection unit. Therefore, the time required to capture an image input to the face detection unit is shortened.

然而，在这样根据规定的焦点信息所摄像的图像中，存在焦点没有对准图像中的被摄物体的可能性。因此，在被输入到脸检测单元中的图像中，存在本来应被检测的脸的图像模糊，被不清晰地摄像的可能性。However, in an image captured based on such predetermined focus information, the subject in the image may not be in focus. Therefore, in the image input to the face detection means, there is a possibility that the image of the face to be detected is blurred and captured indistinctly.

因此，本发明的第一方式包含的脸检测单元构成为也检测被不清晰地摄像的脸的图像。具体地说，脸检测单元根据因人脸的凹凸或器官而产生的多个特征区域的统计量的相对值检测人脸。即，本发明的第一方式包含的脸检测单元不是根据由图像中的边缘等构成的形状，而是根据特征区域的统计量的相对值来进行脸的检测。特征区域的统计量是特征区域内的像素值的合计、平均、分布等的值。相对值是例如比、差分、统计距离、互熵等的值。并且，也可以根据特征区域的浓淡值的相对值来执行脸检测。Therefore, the face detection means included in the first aspect of the present invention is configured to also detect an image of a face that is captured indistinctly. Specifically, the face detection unit detects a human face based on relative values of statistics of a plurality of feature regions due to unevenness or organs of the human face. That is, the face detection unit included in the first aspect of the present invention detects faces not based on shapes formed of edges in an image, but based on relative values of statistical quantities of feature regions. The statistics of the feature area are values such as the sum, average, distribution, etc. of the pixel values in the feature area. Relative values are values such as ratios, differences, statistical distances, cross-entropy, and the like. Also, face detection may be performed based on relative values of gradation values of feature regions.

在被不清晰地摄像的图像中，难以识别由边缘等构成的形状。然而，关于某特征区域的统计量，即使是例如被不清晰地摄像的图像也能取得。因此，本发明的第一方式包含的脸检测单元可以从根据规定的焦点信息所摄像的图像(处理对象的图像)中检测人脸。In an image captured indistinctly, it is difficult to recognize a shape composed of edges or the like. However, statistics on a certain characteristic region can be acquired even in an image that is not clearly captured, for example. Therefore, the face detection means included in the first aspect of the present invention can detect a human face from an image captured based on predetermined focus information (an image to be processed).

决定单元根据由脸检测单元所检测出的脸，决定在实施摄像时进行对焦的被摄物体和/或成为进行曝光控制的对象的被摄物体。在实施摄像时进行对焦的被摄物体和成为进行曝光控制的对象的被摄物体可以是由脸检测单元所检测出的脸本身，也可以是构成所检测出的脸的各部位(眼、鼻、口、耳、额等)。The determination unit determines a subject to be focused on during imaging and/or a subject to be subjected to exposure control based on the face detected by the face detection unit. The subject to be focused on during imaging and the subject to be subject to exposure control may be the face itself detected by the face detection unit, or the parts (eyes, nose, etc.) constituting the detected face. , mouth, ear, forehead, etc.).

在这样构成的本发明的第一方式中的对象决定装置中，从根据任意的焦点信息(规定的焦点信息)所摄像的图像中检测人脸，根据所检测出的脸决定成为对焦对象的被摄物体或成为进行曝光控制的对象的被摄物体。如上所述，通过预先按照规定的焦点信息对摄像装置进行焦点控制，可快速地摄像根据规定的焦点信息来摄像的图像。并且，从该图像中检测人脸，决定成为对焦对象的被摄物体或成为进行曝光控制的对象的被摄物体。In the object determining device in the first aspect of the present invention thus constituted, a human face is detected from an image captured based on arbitrary focus information (predetermined focus information), and a subject to be focused is determined based on the detected face. subject or a subject subject to exposure control. As described above, by performing focus control on the imaging device in advance according to predetermined focus information, an image captured based on predetermined focus information can be quickly captured. Then, a human face is detected from the image, and a subject to be focused on or a subject to be subjected to exposure control is determined.

因此，根据本发明的第一方式中的对象决定装置，在摄像装置中在预备摄像阶段不进行焦点控制，可快速地检测和决定成为进行对焦或曝光控制的对象的主要被摄物体(脸或脸的部位等)。Therefore, according to the object determination device in the first aspect of the present invention, the imaging device does not perform focus control in the preliminary imaging stage, and can quickly detect and determine the main subject (face or subject) to be subjected to focus or exposure control. parts of the face, etc.).

并且，根据本发明的第一方式中的对象决定装置，即使在构成为在摄像装置中取得多张预备摄像的图像的情况下，也可以从更少的预备摄像图像中检测脸。因此，可削减预备摄像所需要的时间和脸检测所需要的时间。Furthermore, according to the object determining device in the first aspect of the present invention, even when the imaging device is configured to acquire a plurality of pre-captured images, faces can be detected from fewer pre-captured images. Therefore, the time required for preliminary imaging and the time required for face detection can be reduced.

本发明的第一方式中的脸检测单元可以构成为按以下所述来检测人脸。即，首先在所摄像的图像中决定关注区域。关注区域是与所摄像的图像大小相同的区域或者更小的区域。The face detection unit in the first aspect of the present invention may be configured to detect human faces as follows. That is, first, a region of interest is determined in a captured image. The attention area is an area that is the same size as the captured image or smaller.

然后，脸检测单元求出该关注区域内的特征区域即第一区域和第二区域的统计量的相对值。第一区域和第二区域分别是关注区域内的不同区域。Then, the face detection unit calculates the relative value of the statistics of the first region and the second region which are characteristic regions within the region of interest. The first area and the second area are respectively different areas within the area of interest.

然后，脸检测单元根据所求出的统计量的相对值，判断在关注区域内是否包含有人脸。脸检测单元可以构成为通过例如对某阈值和统计量的相对值进行比较来进行该判断。并且，脸检测单元也可以构成为通过参照将统计量的相对值、和在算出某统计量的相对值的情况下表示在关注区域内是否包含有人脸的值对应起来进行存储的表，来进行该判断。而且，脸检测单元根据该判断来检测人脸。Then, the face detection unit judges whether or not a human face is included in the region of interest based on the obtained relative value of the statistic. The face detection unit may be configured to make this determination by, for example, comparing a certain threshold value with a relative value of a statistic. In addition, the face detection unit may be configured to perform the operation by referring to a table that stores relative values of statistics in association with values indicating whether or not a human face is included in the region of interest when a relative value of a certain statistics is calculated. The judgment. Also, the face detection unit detects a human face based on this determination.

本发明的第一方式中的决定单元也可以构成为在脸检测单元检测出多个脸的情况下，按以下所述来决定进行对焦的被摄物体或成为进行曝光控制的对象的被摄物体。即，决定单元首先根据由脸检测单元所检测出的多个脸各自的图像信息决定一个脸。脸的图像信息表示从脸的图像中所获得的信息，例如是脸的朝向、脸的图像的大小、图像内的脸的位置、周围的脸的密度等。The determining unit in the first aspect of the present invention may be configured to determine the subject to be focused on or the subject to be subjected to exposure control when a plurality of faces are detected by the face detection unit as follows: . That is, the determining unit first determines one face based on the respective image information of a plurality of faces detected by the face detecting unit. The image information of a face represents information obtained from an image of a face, such as the orientation of the face, the size of the image of the face, the position of the face in the image, the density of the surrounding faces, and the like.

周围的脸的密度由例如在所检测出的脸的一定周围(一定的周围区域)存在的脸的数目来表示。如果该值大，则在周围存在许多脸，例如表示所检测出的脸的人被包含在集体内。另一方面，如果该值小，则在周围存在的脸少，例如表示在所检测出的脸的人与集体分离的场所，存在单独的或几个人。The density of surrounding faces is represented by, for example, the number of faces present in a certain surrounding area (a certain surrounding area) of a detected face. If this value is large, there are many faces around, for example, a person representing a detected face is included in the group. On the other hand, if the value is small, there are few faces around, which means, for example, that a single person or a few people exist in a place where the detected faces are separated from the group.

而且，决定单元根据所决定的脸，决定进行对焦的被摄物体和/或成为进行曝光控制的对象的被摄物体。Further, the determination unit determines a subject to be focused on and/or a subject to be subjected to exposure control based on the determined face.

在这样构成的本发明的第一方式中，根据多个脸的图像信息，判断根据哪个脸来决定进行对焦的被摄物体和/或成为进行曝光控制的对象的被摄物体。因此，即使在由脸检测单元检测出多个脸的情况下，用户也不用实施选择，而由决定单元从多个脸中选择一个脸。In the first aspect of the present invention thus constituted, it is determined which face is used to determine the subject to be focused on and/or the subject to be subjected to exposure control based on image information of a plurality of faces. Therefore, even when a plurality of faces are detected by the face detection unit, the user does not need to make a selection, and the determination unit selects one face from among the plurality of faces.

本发明的第一方式中的决定单元可以构成为对多个脸各自的图像大小进行比较，根据其大小为最大的脸来决定进行对焦的被摄物体和/或成为进行曝光控制的对象的被摄物体。The determining unit in the first aspect of the present invention may be configured to compare the respective image sizes of a plurality of faces, and determine the subject to be focused on and/or the subject to be subjected to exposure control based on the face whose size is the largest. photograph objects.

多个脸中的最大图像的脸一般在距摄像装置最近的位置处。因此，根据这样构成的本发明的第一方式，用户不用实施脸的选择，而是位于距摄像装置最近的位置处的脸被确定，根据该脸来决定进行对焦的被摄物体或成为进行曝光控制的对象的被摄物体。The face of the largest image among the faces is generally at the closest position to the camera. Therefore, according to the first aspect of the present invention constituted in this way, the user does not need to select a face, but the face at the closest position to the imaging device is specified, and the subject to be focused on or the subject to be exposed is determined based on the face. The subject of the controlled object.

由于对摄像装置而言越是位于跟前的物体，图像就越模糊，因而一般来说必须针对对摄像装置而言位于跟前的物体进行对焦。并且，对摄像装置而言位于跟前的脸与位于后面的脸相比，包含对进行摄像的用户而言的主要被摄物体或者就是主要被摄物体本身的可能性(以下表现为“脸的重要性”)高。在这点上，根据位于距摄像装置最近的位置的脸来决定进行对焦的被摄物体和/或成为进行曝光控制的对象的被摄物体是有效的。Since the image is blurred the closer the object is to the imaging device, it is generally necessary to focus on the object located nearer to the imaging device. In addition, compared with the face located in front of the imaging device, the face located in the back includes the main subject for the user who takes the image or is the possibility of the main subject itself (hereinafter referred to as "the importance of the face"). sex") high. In this regard, it is effective to determine a subject to be focused on and/or a subject to be subjected to exposure control based on the face located closest to the imaging device.

本发明的第一方式中的决定单元也可以构成为针对各脸，根据脸的朝向、脸的大小、在图像内的位置、或者周围的脸的密度中的任意一项或多项而算出得分。此时，决定单元把该算出的得分用作图像信息。The determining unit in the first aspect of the present invention may be configured to calculate a score for each face based on any one or more of the orientation of the face, the size of the face, the position within the image, or the density of surrounding faces. . At this time, the determining unit uses the calculated score as image information.

在这样构成的本发明的第一方式中的对象决定装置中，根据脸的朝向、脸的大小、在图像内的位置、或者周围的脸的密度等来评价各脸。例如，脸的朝向是面向正面的脸被评价为其脸的重要性比面向侧面的脸高。并且，例如，脸的大小较大的脸被评价为脸的重要性比小的脸高。并且，例如，在图像内的位置是接近中央的脸被评价为脸的重要性比位于远离中央的区域内的脸高。并且，周围的脸的密度高的脸被评价为脸的重要性比周围的脸的密度低的脸高。然而，该评价的高低可以由设计者或用户来任意设定。In the object determination device according to the first aspect of the present invention configured in this way, each face is evaluated based on the orientation of the face, the size of the face, the position within the image, or the density of surrounding faces. For example, a face whose face is oriented frontally is evaluated as being more important than a face oriented sideways. And, for example, a face with a large face size is evaluated as having a higher importance of the face than a small face. Also, for example, a face whose position is closer to the center within the image is evaluated as having a higher importance of the face than a face located in a region farther from the center. In addition, a face with a high density of surrounding faces is evaluated as having a higher importance of the face than a face with a low density of surrounding faces. However, the level of this evaluation can be set arbitrarily by the designer or the user.

根据这样构成的本发明的第一方式，根据这种评价的高低设定，选择得分高即重要性高的脸。According to the first aspect of the present invention thus constituted, a face with a high score, that is, a face with high importance is selected based on the level setting of the evaluation.

本发明的第一方式中的决定单元也可以构成为在脸检测单元检测出多个人脸的情况下，按以下所述来决定进行对焦的被摄物体和/或成为曝光控制的对象的被摄物体。即，决定单元按照对由脸检测单元所检测出的多个脸的实物分别进行测距所得的结果，决定进行对焦的被摄物体和/或成为进行曝光控制的对象的被摄物体。该测距可以由例如摄像装置中所具备的测距单元或测距装置来实施，也可以由对象决定装置中所具备的测距单元来实施。然而，该测距由于必须测量摄像装置和脸的实物的距离，因而如果摄像装置和对象决定装置相分离，则不能进行由对象决定装置中所具备的测距单元所实施的测距。The determination unit in the first aspect of the present invention may be configured to determine the subject to be focused on and/or the subject to be subject to exposure control when the face detection unit detects a plurality of faces. object. That is, the determination unit determines the subject to be focused on and/or the subject to be subjected to exposure control according to the results of distance measurement of the faces detected by the face detection unit. This distance measurement may be performed by, for example, a distance measurement unit or a distance measurement device included in the imaging device, or may be performed by a distance measurement unit included in the object determination device. However, since this distance measurement requires measuring the distance between the imaging device and the actual face, if the imaging device and the object determination device are separated, distance measurement cannot be performed by a distance measuring unit included in the object determination device.

测距通过例如对脸的实物射出红外线来实现(有源方式)。并且，测距也可以应用这种有源方式以外的方式(例如无源方式)。而且，决定单元根据其距离为最小的脸来决定进行对焦的被摄物体和/或成为进行曝光控制的对象的被摄物体。这样，根据摄像装置和脸的实物的距离为最小的脸来决定进行对焦的被摄物体或成为进行曝光控制的对象的被摄物体的效果如上所述。Distance measurement is realized by, for example, emitting infrared rays to a real object in the face (active method). In addition, a method other than this active method (for example, a passive method) may be applied to distance measurement. Further, the determination unit determines a subject to be focused on and/or a subject to be subjected to exposure control based on the face whose distance is the smallest. In this way, the effect of determining the subject to be focused on or the subject to be subjected to exposure control based on the face whose distance between the imaging device and the actual face is the smallest is as described above.

〔第二方式〕[Second method]

本发明的第二方式是一种摄像装置，包括摄像单元、脸检测单元、决定单元、信息取得单元以及图像存储单元。A second aspect of the present invention is an imaging device including an imaging unit, a face detection unit, a determination unit, an information acquisition unit, and an image storage unit.

摄像单元摄像基于规定的焦点信息或者由信息取得单元所取得的焦点信息和/或曝光控制信息的图像。规定的焦点信息是预先决定为一个的焦点信息，而不是动态变化的焦点信息。另一方面，由信息取得单元所取得的焦点信息不是预先决定的焦点信息，而是动态变化的焦点信息。更具体地说，是通过脸检测单元、决定单元以及信息取得单元所决定的焦点信息。即，是用于根据所检测出的脸来决定进行对焦的被摄物体，对该所决定的被摄物体对焦的焦点信息，The imaging unit captures an image based on predetermined focus information or focus information and/or exposure control information acquired by the information acquisition unit. The predetermined focus information is predetermined focus information, not dynamically changing focus information. On the other hand, the focus information acquired by the information acquiring means is not predetermined focus information but dynamically changing focus information. More specifically, it is the focus information determined by the face detection unit, determination unit, and information acquisition unit. That is, it is focus information for determining a subject to focus on based on a detected face and focusing on the determined subject,

在没有通过信息取得单元输出的焦点信息的情况下，摄像单元被控制成可取得基于规定的焦点信息的图像的状态。而且，当有摄像指示时，取得基于规定的焦点信息的图像。另一方面，在通过信息取得单元输出有焦点信息或曝光控制信息的情况下，摄像单元被控制成可取得基于这些信息的图像的状态。而且，当有摄像指示时，摄像单元取得基于该焦点信息或曝光控制信息的图像。When there is no focus information output by the information acquisition unit, the imaging unit is controlled so that an image based on predetermined focus information can be acquired. Furthermore, when there is an imaging instruction, an image based on predetermined focus information is acquired. On the other hand, when focus information or exposure control information is output by the information acquisition means, the imaging means is controlled so that an image based on these information can be acquired. Then, when there is an imaging instruction, the imaging unit acquires an image based on the focus information or exposure control information.

脸检测单元从基于规定的焦点信息的处理对象的图像中，根据因人脸的凹凸或器官而产生的多个特征区域的统计量的相对值来检测人脸。决定单元根据由脸检测单元所检测出的脸，决定在实施摄像时进行对焦的被摄物体和/或成为进行曝光控制的对象的被摄物体。本发明的第二方式的脸检测单元和决定单元基本上具有与本发明的第一方式中的脸检测单元和决定单元相同的结构和作用。The face detection unit detects a human face based on relative values of statistics of a plurality of characteristic regions generated by unevenness or organs of the human face from an image of the processing target based on predetermined focus information. The determination unit determines a subject to be focused on during imaging and/or a subject to be subjected to exposure control based on the face detected by the face detection unit. The face detection unit and decision unit of the second aspect of the present invention basically have the same configuration and function as those of the face detection unit and decision unit of the first aspect of the present invention.

信息取得单元取得用于对由决定单元所决定的被摄物体对焦的焦点信息和/或用于把由决定单元所决定的被摄物体作为对象进行曝光控制的曝光控制信息。The information acquiring unit acquires focus information for focusing on the subject determined by the determining unit and/or exposure control information for performing exposure control on the subject determined by the determining unit.

图像存储单元存储由摄像单元根据由信息取得单元所取得的焦点信息和/或曝光控制信息而摄像的图像。即，当由信息取得单元取得了焦点信息或曝光控制信息时，摄像单元根据该焦点信息或曝光控制信息实施焦点控制或曝光控制。通过该焦点控制，对由决定单元所决定的被摄物体进行对焦。并且，通过该曝光控制，实施与由决定单元所决定的被摄物体对应的曝光控制。然后，摄像单元执行摄像，图像存储单元存储所摄像的图像的数据。The image storage unit stores an image captured by the imaging unit based on the focus information and/or exposure control information acquired by the information acquisition unit. That is, when the focus information or exposure control information is acquired by the information acquisition unit, the imaging unit performs focus control or exposure control based on the focus information or exposure control information. Through this focus control, the subject determined by the determination unit is focused. And, through this exposure control, exposure control corresponding to the subject determined by the determination unit is performed. Then, the imaging unit performs imaging, and the image storage unit stores data of the imaged image.

并且，本发明的第二方式可以构成为还具有测距单元，其对从摄像装置到任意被摄物体的距离进行测距。然而，在信息取得单元构成为通过对被摄物体的实物和摄像装置的距离进行测距来取得焦点信息的情况下，可以具备信息取得单元作为测距单元的替代单元。即，在该情况下，信息取得单元作为测距单元进行动作。Furthermore, the second aspect of the present invention may be configured to further include a distance measuring unit that measures the distance from the imaging device to an arbitrary subject. However, when the information obtaining unit is configured to obtain focus information by measuring the distance between the actual subject and the imaging device, the information obtaining unit may be provided as a substitute for the distance measuring unit. That is, in this case, the information acquiring unit operates as a distance measuring unit.

在这样构成本发明的第二方式的情况下，决定单元构成为在脸检测单元检测出多个人脸的情况下，按照测距单元对多个脸的实物分别进行测距所得的结果，根据其距离为最小的脸来决定进行对焦的被摄物体和/或成为进行曝光控制的对象的被摄物体。In the case where the second aspect of the present invention is constituted in this way, the determining unit is configured so that when the face detecting unit detects a plurality of human faces, the distance measuring unit performs distance measurement on the actual objects of the plurality of faces according to the results obtained, and based on the result A subject to be focused on and/or a subject to be subject to exposure control is determined based on the face with the smallest distance.

在本发明的第二方式中，在摄像装置中配备作为本发明的第一方式的对象决定装置。因此，作为摄像装置，可获得与本发明的第一方式相同的效果。In a second aspect of the present invention, the imaging device is equipped with the object determination device of the first aspect of the present invention. Therefore, as an imaging device, the same effect as that of the first aspect of the present invention can be obtained.

〔第三方式〕[Third method]

本发明的第三方式是一种对象决定方法，是包含以下各步骤的方法。首先，信息处理装置从根据任意焦点信息所摄像的图像中，根据因人脸的凹凸或器官而产生的多个特征区域的统计量的相对值来检测人脸。然后，信息处理装置根据所检测出的脸，决定在实施摄像时进行对焦的被摄物体和/或成为进行曝光控制的对象的被摄物体。这样构成的本发明的第三方式获得与本发明的第一方式相同的效果。A third aspect of the present invention is an object determination method including the following steps. First, the information processing device detects a human face based on relative values of statistics of a plurality of characteristic regions generated by unevenness or organs of the human face from an image captured based on arbitrary focus information. Then, based on the detected face, the information processing device determines a subject to be focused on when imaging is performed and/or a subject to be subjected to exposure control. The third aspect of the present invention constituted in this way achieves the same effects as those of the first aspect of the present invention.

〔第四方式〕[Fourth method]

本发明的第四方式是一种摄像方法，是包含以下各步骤的方法。以下，列举该各步骤。信息处理装置指示摄像装置对基于规定的焦点信息或者由信息取得单元所取得的焦点信息的图像进行摄像的步骤。信息处理装置从由摄像装置根据规定的焦点信息所摄像的图像中，根据因人脸的凹凸或器官而产生的多个特征区域的统计量的相对值来检测人脸的检测步骤。信息处理装置根据所检测出的脸，决定在实施摄像时进行对焦的被摄物体和/或成为进行曝光控制的对象的被摄物体的决定步骤。信息处理装置取得用于对所决定的被摄物体进行对焦的焦点信息和/或用于把所决定的被摄物体作为对象来进行曝光控制的曝光控制信息的步骤。信息处理装置指示存储装置存储由摄像装置根据所取得的焦点信息和/或曝光控制信息所摄像的图像的步骤。A fourth aspect of the present invention is an imaging method including the following steps. Each of these steps is listed below. A step in which the information processing device instructs the imaging device to capture an image based on predetermined focus information or focus information acquired by an information acquisition unit. The information processing device detects a human face based on relative values of statistical quantities of a plurality of characteristic regions generated by unevenness or organs of the human face from an image captured by the imaging device based on predetermined focus information. The information processing device determines a subject to focus on when performing imaging and/or a subject to be subject to exposure control based on the detected face. A step in which the information processing device acquires focus information for focusing on the determined subject and/or exposure control information for performing exposure control with the determined subject as a target. The information processing device instructs the storage device to store the image captured by the imaging device based on the acquired focus information and/or exposure control information.

并且，本发明的第四方式可以构成为还包含信息处理装置指示测距装置对从摄像装置到任意被摄物体的距离进行测距的步骤。在该情况下，如果在检测步骤中检测出多个人脸，则信息处理装置在决定步骤中，按照测距装置对多个脸的实物分别进行测距所得的结果，根据其距离为最小的脸来决定进行前述对焦的被摄物体和/或成为进行曝光控制的对象的被摄物体。Furthermore, the fourth aspect of the present invention may be configured to further include a step in which the information processing device instructs the distance measuring device to measure the distance from the imaging device to an arbitrary subject. In this case, if a plurality of human faces are detected in the detection step, the information processing device, in the determination step, according to the results obtained by the ranging device for the actual objects of the plurality of faces respectively, according to the face whose distance is the smallest The subject to be focused on and/or the subject to be subject to exposure control is determined.

在这样构成的本发明的第四方式中，把本发明的第三方式中的对象决定方法作为包含在摄像方法中的步骤来执行。因此，在本发明的第四方式中，可取得与本发明的第三方式相同的效果。In the fourth aspect of the present invention thus constituted, the object determination method in the third aspect of the present invention is executed as a step included in the imaging method. Therefore, in the fourth aspect of the present invention, the same effects as those of the third aspect of the present invention can be obtained.

〔第五方式〕[fifth way]

本发明的第五方式是一种对象决定装置，包括检测单元和决定单元。检测单元从根据任意的焦点信息所摄像的图像中检测具有人的肤色的被摄物体。作为具有人的肤色的被摄物体的例子，有人的脸、手、身体等。通过这样构成检测单元，尽管不能准确地检测人物的脸，然而可削减检测处理所需要的时间。并且，也能削减检测单元的安装所需要的时间和成本。A fifth aspect of the present invention is an object determination device including detection means and determination means. The detection unit detects a subject having human skin color from an image captured based on arbitrary focus information. As examples of subjects having human skin color, there are human faces, hands, bodies, and the like. By configuring the detection means in this way, although the face of a person cannot be accurately detected, the time required for the detection process can be reduced. In addition, the time and cost required for mounting the detection unit can also be reduced.

决定单元根据由检测单元所检测出的被摄物体，决定在实施摄像时进行对焦的被摄物体和/或成为进行曝光控制的对象的被摄物体。The determination unit determines a subject to be focused on when imaging is performed and/or a subject to be subjected to exposure control based on the subject detected by the detection unit.

在第五方式中，检测单元检测具有人的肤色的被摄物体。即，本发明的第五方式所包括的检测单元与本发明的第一方式一样，不进行基于由图像中的边缘等构成的形状的检测。In a fifth aspect, the detection unit detects a subject having human skin color. That is, like the first aspect of the present invention, the detection means included in the fifth aspect of the present invention does not perform detection based on shapes such as edges in the image.

在被不清晰地摄像的图像中，难以识别由边缘等构成的形状。然而，即使从例如被不清晰地摄像的图像中也能检测出具有某特定颜色的被摄物体。因此，本发明的第五方式所包括的检测单元能够从根据规定的焦点信息所摄像的图像中检测存在是人脸的可能性的被摄物体，即具有人的肤色的被摄物体。In an image captured indistinctly, it is difficult to recognize a shape composed of edges or the like. However, an object having a certain color can be detected even from, for example, an image that is not clearly captured. Therefore, the detection unit included in the fifth aspect of the present invention can detect a subject that may be a human face, that is, a subject with human skin color, from an image captured based on predetermined focus information.

〔第一方式的变形〕[Variation of the first form]

本发明的第一方式可以构成为还包括帧取得单元，该帧取得单元从根据任意的焦点信息所摄像的包含多个帧的时序图像中取得某一帧的图像作为处理对象的图像。在这样构成的情况下，第一方式中的脸检测单元把由帧取得单元所取得的某一帧的图像作为处理对象。The first aspect of the present invention may be configured to further include a frame acquisition unit that acquires an image of a certain frame as an image to be processed from time-series images including a plurality of frames captured based on arbitrary focus information. In such a configuration, the face detection unit in the first aspect takes an image of a certain frame acquired by the frame acquisition unit as a processing target.

在这样构成的本发明的第一方式中，对于包含多个帧的时序图像，可取得与本发明的第一方式相同的作用/效果。In the first aspect of the present invention thus constituted, the same operation/effect as that of the first aspect of the present invention can be obtained for time-series images including a plurality of frames.

并且，本发明的第一方式中的决定单元可以构成为在由脸检测单元检测出多个人脸时，根据该多个人脸的位置，从该多个人脸中决定在实施摄像时成为进行对焦的对象的脸和/或成为进行曝光控制的对象的脸。通过这样构成，可取得与后述的本发明的第六方式相同的作用/效果。In addition, the determining unit in the first aspect of the present invention may be configured to determine, from among the plurality of faces, which faces to focus on when imaging is performed, based on the positions of the plurality of faces when the face detection unit detects a plurality of faces. The subject's face and/or the subject's face subject to exposure control. With such a configuration, the same operations and effects as those of the sixth aspect of the present invention described later can be obtained.

并且，本发明的第一方式中的决定单元可以构成为在由脸检测单元检测出多个人脸时，以所检测出的脸的数目为基准把位于其正中的人脸决定为在实施摄像时成为进行对焦的对象的脸和/或成为进行曝光控制的对象的脸。通过这样构成，可取得与后述的本发明的第七方式相同的作用/效果。In addition, the determining unit in the first aspect of the present invention may be configured to determine, based on the number of detected faces when a plurality of human faces are detected by the face detecting unit, the human face located in the center as the face at the time of imaging. A face to be focused on and/or a face to be controlled for exposure. With such a configuration, the same operations and effects as those of the seventh aspect of the present invention described later can be obtained.

并且，本发明的第一方式中的决定单元可以构成为包括：分类单元，在由脸检测单元检测出多个人脸时，其把所检测出的多个人脸分类为多个集合；暂定决定单元，其针对多个集合中的各个集合，从该集合内所包含的人脸中决定在实施摄像时成为进行对焦的对象的脸和/或成为进行曝光控制的对象的脸；以及最终决定单元，其从由暂定决定单元所决定的脸中最终决定成为对象的脸。通过这样构成，可取得与后述的本发明的第八方式相同的作用/效果。Furthermore, the determining unit in the first aspect of the present invention may be configured to include: a classifying unit that, when a plurality of human faces is detected by the face detecting unit, classifies the detected plurality of human faces into a plurality of sets; a unit for determining, for each of the plurality of sets, a face to be focused upon and/or a face to be subjected to exposure control from among faces included in the set; and a final determination unit , which finally determines the face to be the subject from among the faces determined by the tentative determination unit. With such a configuration, the same operations and effects as those of the eighth aspect of the present invention described later can be obtained.

并且，本发明的第一方式可以构成为还包含区决定单元，其根据脸检测单元的检测结果来决定选择区，该选择区用于决定在实施摄像时成为进行对焦的对象的脸和/或成为进行曝光控制的对象的脸。在这样构成的情况下，决定单元构成为从选择区内所包含的脸中决定成为对象的脸。通过这样构成，可取得与后述的本发明的第九方式相同的作用/效果。Furthermore, the first aspect of the present invention may be configured to further include an area determination unit that determines a selection area for determining a face and/or a face to be focused on when imaging is performed based on a detection result of the face detection unit. The face that becomes the subject of exposure control. In such a configuration, the determining unit is configured to determine a target face from faces included in the selection area. With such a configuration, the same operations and effects as those of the ninth aspect of the present invention described later can be obtained.

并且，本发明的第一方式可以构成为还包含：判断单元，在由脸检测单元检测出多个人脸时，其判断所检测出的人脸中的最大的脸；以及选择单元，其从所检测出的脸中选择最大的脸和具有以该最大的脸的大小为基准的规定范围内的大小的脸。在这样构成的情况下，决定单元构成为从所选择的脸中决定在实施摄像时成为进行对焦的对象的脸和/或成为进行曝光控制的对象的脸。通过这样构成，可取得与后述的本发明的第十方式相同的作用/效果。Furthermore, the first aspect of the present invention may be configured to further include: a judging unit that judges the largest face among the detected human faces when a plurality of human faces are detected by the face detection unit; Among the detected faces, the largest face and a face having a size within a predetermined range based on the size of the largest face are selected. In such a configuration, the determining unit is configured to determine, from among the selected faces, a face to be focused on when imaging is performed and/or a face to be subjected to exposure control. With such a configuration, the same operations and effects as those of the tenth aspect of the present invention described later can be obtained.

并且，本发明的第一方式可以构成为还包括：分类单元，在由前述脸检测单元检测出多个人脸时，将所检测出的多个人脸根据各自的位置分类为多个集合；以及集合决定单元，其从多个集合中决定选择集合，该选择集合用于决定在实施摄像时成为进行对焦的对象的脸和/或成为进行曝光控制的对象的脸。在这样构成的情况下，决定单元构成为从由集合决定单元所决定的选择集合内所包含的脸中决定成为对象的脸。In addition, the first aspect of the present invention may be configured to further include: a classifying unit that, when a plurality of human faces is detected by the aforementioned face detection unit, classifies the detected plurality of human faces into a plurality of sets according to their respective positions; and the set A determining unit that determines a selected set for determining a face to be focused on and/or a face to be controlled for exposure, from a plurality of sets. In such a configuration, the determining unit is configured to determine a target face from faces included in the selection set determined by the set determining unit.

〔第二方式的变形〕[Variation of the second method]

本发明的第二方式可以构成为摄像单元摄像包含多个帧的时序图像。本发明的第二方式在这样构成的情况下，构成为还包括帧取得单元，其从由摄像单元所摄像的包含多个帧的时序图像中取得某一帧的图像。并且，脸检测单元构成为从由帧取得单元所取得的图像中检测人脸。并且，图像存储单元构成为存储由摄像单元所摄像的时序图像。A second aspect of the present invention may be configured such that the imaging unit captures time-series images including a plurality of frames. When the second aspect of the present invention is configured in this way, it is configured to further include a frame acquisition unit that acquires an image of a certain frame from time-series images including a plurality of frames captured by the imaging unit. Furthermore, the face detection unit is configured to detect a human face from the image acquired by the frame acquisition unit. Furthermore, the image storage unit is configured to store time-series images captured by the imaging unit.

在这样构成的本发明的第二方式中，对于包含多个帧的时序图像，可取得与本发明的第二方式相同的作用/效果。In the second aspect of the present invention thus constituted, the same operations/effects as those of the second aspect of the present invention can be obtained for time-series images including a plurality of frames.

并且，本发明的第二方式可以构成为还包括显示单元，其把由决定单元所决定的人脸与其它脸区别开来显示。显示单元例如把框显示在所决定的人脸的周围，从而与其它脸区别开来显示。另外，显示单元例如在所决定的人脸的周围显示与在其它脸的周围所显示的框有不同的颜色、粗细或形状的框，从而与其它脸区别开来显示。另外，显示单元例如对所决定的人脸实施与其它脸不同的图像处理并显示，从而与其它脸区别开来显示。显示单元通过进行这种显示，使用户可以容易地判断是根据哪个脸来决定成为对焦对象的被摄物体等的。Furthermore, the second aspect of the present invention may be configured to further include display means for displaying the human face determined by the determination means in distinction from other faces. The display unit displays, for example, a frame around the determined face to distinguish it from other faces. In addition, the display unit displays, for example, a frame having a different color, thickness, or shape around the determined face from frames displayed around other faces, thereby distinguishing the frame from other faces. In addition, the display unit performs, for example, image processing different from other faces on the determined face and displays it so as to be distinguished from other faces. By performing such a display, the display unit enables the user to easily determine which face is used to determine the subject to be focused on, and the like.

〔第六方式〕[Sixth way]

为了达到上述第二目的，本发明采用以下结构。本发明的第六方式是一种对象决定装置，包括检测单元和决定单元。本发明的第六方式根据所检测的脸的位置，决定成为应进行对焦和曝光控制的对象的脸。另外，由本发明的第六方式所决定的脸可以作为任何处理的对象，而不限于对焦和曝光控制。例如，可以作为进行颜色校正(白平衡校正)、轮廓强调等处理的对象。In order to achieve the above-mentioned second object, the present invention employs the following structures. A sixth aspect of the present invention is an object determination device including detection means and determination means. In the sixth aspect of the present invention, a face to be subjected to focus and exposure control is determined based on the position of the detected face. In addition, the face determined by the sixth aspect of the present invention may be the target of any processing, not limited to focus and exposure control. For example, it can be used as a target for color correction (white balance correction), outline enhancement, and the like.

具体地说，检测单元从所输入的图像中检测人脸。检测单元只要是能从图像中检测多个人脸的技术，可以应用现有的任何脸检测技术。例如，可以检测具有肤色的区域作为脸区域。作为其它例子，有使用脸或脸的部位的模板的检测技术、以及根据浓淡差来检测脸区域的技术等。Specifically, the detection unit detects a human face from an input image. Any existing face detection technology may be applied to the detection unit as long as it is a technology capable of detecting a plurality of human faces from an image. For example, an area with skin color can be detected as a face area. As other examples, there are detection techniques using a template of a face or a part of a face, and techniques for detecting a face region based on a difference in gradation.

在检测单元检测出多个人脸时，决定单元根据该多个人脸的位置，从该多个人脸中决定在实施摄像时成为进行对焦的对象的脸和/或成为进行曝光控制的对象的脸。决定单元不是通过独立考虑各个脸的位置来进行决定，而是根据多个脸的位置来进行决定。即，决定单元至少根据多个脸的相对位置关系来进行决定。决定单元可以把脸的位置看作脸部所包含的点的位置，也可以看作脸区域的位置。When the detecting unit detects a plurality of human faces, the determining unit determines a face to be focused on and/or a face to be subjected to exposure control during imaging from among the plurality of faces based on the positions of the plurality of faces. The decision unit does not make a decision by considering the position of each face independently, but makes a decision based on the positions of a plurality of faces. That is, the determining unit determines based on at least the relative positional relationship of the plurality of faces. The decision unit may regard the position of the face as the position of the points contained in the face, or as the position of the face region.

另外，可以构成为对由决定单元所决定的脸的一部分(例如眼、鼻、额、口、耳)进行对焦或曝光控制。In addition, it may be configured to perform focus or exposure control on a part of the face (for example, eyes, nose, forehead, mouth, and ears) determined by the determination unit.

根据本发明的第六方式，根据所检测出的多个脸的位置，决定成为对焦或曝光控制的对象的脸。因此，即使在从进行了预备摄像得到的图像中检测出比应成为对焦或曝光控制的对象的脸的数目多的脸的情况下，也能决定规定数目的脸。According to the sixth aspect of the present invention, based on the positions of the detected faces, the face to be controlled for focus or exposure is determined. Therefore, even when more faces than faces to be targeted for focus or exposure control are detected from the image obtained by performing preliminary imaging, the predetermined number of faces can be determined.

并且，不是针对所检测出的各个脸的位置独立考虑，而是根据多个脸的相对位置决定成为对焦或曝光控制的对象的脸。因此，根据所检测出的多个脸的状况(例如脸的密集程度、脸密集的位置)来决定成为对焦或曝光控制的对象的脸。因此，能够进行与所检测出的多个脸的状况对应的对焦处理或曝光控制。In addition, instead of considering the position of each detected face independently, the face to be the target of focus or exposure control is determined based on the relative positions of a plurality of faces. Therefore, the face to be the target of focus or exposure control is determined based on the conditions of the detected faces (for example, the degree of crowding of faces, the position where faces are crowded). Therefore, it is possible to perform focus processing or exposure control according to the conditions of the detected faces.

下面，对本发明的第六方式中的决定单元的更具体结构进行说明。本发明的第六方式中的决定单元可以构成为包括中心决定单元和脸决定单元。Next, a more specific configuration of the determining means in the sixth aspect of the present invention will be described. The determination unit in the sixth aspect of the present invention may be configured to include a center determination unit and a face determination unit.

中心决定单元根据在检测单元中检测出的多个人脸的位置，决定这些多个人脸的位置的中心。这里所说的中心表示概念中心。例如，在针对某3点称为中心的情况下，表示外心、内心、垂心、重心等多个概念中心。这样，这里所说的中心表示根据某概念所求出的正中。The center determination unit determines the centers of the positions of the plurality of human faces based on the positions of the plurality of human faces detected by the detection unit. The center mentioned here means the conceptual center. For example, when referring to a certain three points as a center, it means a plurality of concept centers such as circumcenter, incenter, orthocenter, and center of gravity. In this way, the center mentioned here means the median obtained from a certain concept.

具体地说，中心决定单元可以构成为求出与多个人脸的位置外接的多边形的中心。此时，多边形可以是具有预先决定的数目的顶点的多边形。在该情况下，所生成的多边形没有必要一定具有所有的人脸作为顶点。Specifically, the center determination unit may be configured to obtain the centers of polygons circumscribing the positions of the plurality of human faces. At this time, the polygon may be a polygon having a predetermined number of vertices. In this case, the generated polygon does not necessarily have all human faces as vertices.

例如，在外接的多边形是三角形的情况下，中心决定单元取得所求出的三角形的外心、内心、垂心、重心中的任意一个作为中心。For example, when the circumscribed polygon is a triangle, the center determining means acquires any one of the circumcenter, incenter, orthocenter, and center of gravity of the obtained triangle as the center.

并且，例如，在外接的多边形是四边形的情况下，中心决定单元取得所求出的四边形的对角线的交点作为中心。并且，例如，在外接的多边形是四边形的情况下，中心决定单元可以构成为把该四边形分割成二个三角形，分别针对各三角形取得外心、内心、垂心、重心中的任意一个，根据所取得的2点取得中心。Furthermore, for example, when the circumscribing polygon is a quadrangle, the center determining means acquires the intersection point of the obtained diagonals of the quadrangle as the center. And, for example, when the circumscribed polygon is a quadrilateral, the center determination unit may be configured to divide the quadrilateral into two triangles, obtain any one of the circumcenter, incenter, orthocenter, and center of gravity for each triangle, and use the obtained The 2 points get the center.

并且，例如，在外接的多边形的顶点数是大于等于5点的情况下，中心决定单元可以构成为把外接的多边形分割成多个三角形，针对各三角形取得外心、内心、垂心、重心中的任意一个，使用所取得的点构成新的多边形，通过针对该多边形重复上述处理来取得中心。And, for example, when the number of vertices of the circumscribed polygon is equal to or greater than 5 points, the center determination unit may be configured to divide the circumscribed polygon into a plurality of triangles, and obtain the circumcenter, incenter, orthocenter, and center of gravity for each triangle. Either way, a new polygon is constructed using the acquired points, and the center is acquired by repeating the above-mentioned process for the polygon.

并且，中心决定单元可以构成为把多个人脸的位置的重心决定为中心。Furthermore, the center determination unit may be configured to determine the center of gravity of the positions of the plurality of faces as the center.

下面，对脸决定单元的结构进行说明。脸决定单元根据由中心决定单元所求出的中心的位置，决定成为对象的脸。脸决定单元根据在实施摄像时成为对焦对象的脸和/或成为进行曝光控制的对象的脸的数目，决定成为对象的脸。Next, the configuration of the face determination unit will be described. The face determination unit determines a target face based on the position of the center obtained by the center determination unit. The face determination unit determines a target face based on the number of faces to be focused on and/or faces to be subjected to exposure control when imaging is performed.

具体地说，脸决定单元可以构成为把位于距中心最近的人脸决定为成为对象的脸。Specifically, the face determination unit may be configured to determine the face of the person closest to the center as the target face.

并且，脸决定单元可以构成为从位于距中心规定距离内的脸中，按照规定的基准决定成为对象的脸。规定的基准例如可以是基于脸的大小的基准(例如决定最大的脸的基准，决定中间大小的脸的基准，决定最小的脸的基准)，可以是基于图像中的脸的位置的基准(例如距图像中央最近的脸的基准)，可以是基于脸的朝向的基准(例如面向正面的脸的基准)，可以是基于类似于脸的基准(例如决定表示类似于脸的程度的量为最大的脸的基准)，可以是基于根据脸所推测的性别的基准(例如决定被推测为男性的脸的基准，决定被推测为女性的脸的基准)，可以是基于根据脸所推测的年代的基准(决定被推测为最年轻的年代的脸的基准，决定被推测为中间年代的脸的基准)，也可以是其它任何基准。并且，规定的基准也可以是将上述基准适当进行多个组合后的基准。Furthermore, the face determination unit may be configured to determine a target face according to a predetermined criterion from among faces located within a predetermined distance from the center. The predetermined criterion may be, for example, a criterion based on the size of the face (for example, a criterion for determining the largest face, a criterion for determining a middle-sized face, and a criterion for determining the smallest face), or a criterion based on the position of the face in the image (for example, The face closest to the center of the image) may be a reference based on the orientation of the face (for example, a reference for a face facing the front), and may be a reference based on a similar face (for example, the amount indicating the degree of similarity to a face is determined to be the largest face reference), which may be a reference based on the gender estimated from the face (for example, a reference to determine a face estimated to be male, a reference to determine a face estimated to be female), or a reference based on the age estimated from the face (the reference for determining the face estimated to be the youngest age, the reference for determining the face estimated to be the middle age), or any other reference. Furthermore, the predetermined standard may be a combination of a plurality of the above-mentioned standards as appropriate.

在这样构成的本发明的第六方式中，作为多个脸的相对位置的基准，根据多个脸的位置的中心，来决定成为对焦或曝光控制的对象的脸。例如，位于多个脸的大致中心处的脸被决定为应成为对焦或曝光控制的对象的脸。在对集体进行摄影的状况(多个脸密集的状况)下，多数情况用户都是对该集体的中央附近的脸(无意识地)进行对焦或曝光控制。因此，通过自动实现这种用户的手动操作，可节省用户的工夫。因此，在自动聚焦或自动曝光控制中，一般期望的是对位于集体中的多个脸的中央附近的脸进行对焦或曝光控制。而且，根据上述构成的本发明的第六方式，可自动进行上述的期望的控制，可节省用户的工夫。In the sixth aspect of the present invention thus constituted, as a reference of the relative positions of the plurality of faces, the face to be targeted for focus or exposure control is determined based on the center of the positions of the plurality of faces. For example, a face located approximately at the center of a plurality of faces is determined to be a face to be targeted for focus or exposure control. In a situation where a group is photographed (a situation where many faces are crowded), the user often performs focus or exposure control on a face near the center of the group (unconsciously). Therefore, by automatically implementing such a user's manual operation, the user's effort can be saved. Therefore, in automatic focus or automatic exposure control, it is generally desirable to perform focus or exposure control on a face located near the center of a plurality of faces in a group. Furthermore, according to the sixth aspect of the present invention configured as described above, the above-mentioned desired control can be automatically performed, thereby saving the user's effort.

并且，本发明的第六方式中的决定单元可以构成为把多个人脸中的位于最下方的脸决定为成为前述对象的脸。另外，最下方的脸没有必要是严格地位于最下方的脸，可以是例如位于从下开始第2的脸，也可以是位于从下开始第3的脸。在人脸纵向排列的状态下进行集合拍照的情况下，一般越是位于下方的脸是距摄像装置的距离越近的脸。例如，在前排人蹲下、后排人站立进行拍照的情况下，就成为上述状态。并且，例如，在使用高台进行集合拍照的情况下，就成为上述状态。因此，通过把位于最下方的脸视为对焦或曝光控制的对象的脸，对距摄像装置最近的脸执行对焦或曝光控制的可能性变大。In addition, the determining means in the sixth aspect of the present invention may be configured to determine a lowermost face among a plurality of human faces as the aforementioned target face. In addition, the lowermost face does not have to be strictly the lowermost face, and may be, for example, the second face from the bottom or the third face from the bottom. When collective photography is performed with human faces arranged vertically, generally, the lower the face is, the closer the distance to the camera is. For example, when the person in the front row squats down and the person in the back row takes a picture while standing, the above-mentioned state occurs. And, for example, in the case of collective photography using a high platform, it becomes the above-mentioned state. Therefore, by regarding the face located at the bottom as the face to be controlled by focus or exposure, it is more likely that focus or exposure control will be performed on the face closest to the imaging device.

本发明的第六方式可以构成为还包括分类单元和集合决定单元。分类单元在由检测单元检测出多个人脸时，将所检测出的多个人脸根据各自的位置分类为多个集合。更具体地说，分类单元通过使用一般的分类方法，把所检测出的多个脸分类为多个集合，以使得图像中的位置相近的脸被包含在同一集合内。A sixth aspect of the present invention may be configured to further include classification means and aggregation determination means. When the detection unit detects a plurality of human faces, the classification unit classifies the detected plurality of human faces into a plurality of sets according to their respective positions. More specifically, the classifying unit classifies the detected faces into a plurality of sets by using a general classifying method so that faces with close positions in the image are included in the same set.

集合决定单元从由分类单元所分类的多个集合中决定(选择)用于决定成为对焦或曝光控制的对象的脸的集合(即，包含成为对焦或曝光控制的对象的脸的集合：称为“选择集合”)。即，从由集合决定单元所决定的选择集合内所包含的脸中最终决定成为对焦或曝光控制的对象的脸。The set determining unit decides (selects) a set for determining faces to be targeted for focus or exposure control from among a plurality of sets classified by the classifying unit (that is, a set including faces to be targeted for focus or exposure control: called "Select Collection"). That is, the face to be the focus or exposure control target is finally determined from the faces included in the selection set determined by the set determination unit.

集合决定单元根据哪种基准来决定选择集合均可。例如，集合决定单元可以根据各集合内所包含的脸的数目决定选择集合，可以根据各集合内所包含的多个脸的位置的中心决定选择集合，可以根据基于各集合内所包含的脸所取得的特征量(例如，基于脸的大小、脸的朝向、脸的数目等所取得的量，表示根据脸所推测的特定性别的频率、根据脸所推测的特定年代的频率或者类似于脸的程度的量)决定选择集合。The set decision unit may decide the selection set according to any kind of reference. For example, the set determining unit can determine the selection set according to the number of faces contained in each set, can determine the selection set according to the center of the positions of a plurality of faces contained in each set, and can determine the selection set based on the number of faces contained in each set. The acquired feature quantity (for example, the quantity acquired based on the size of the face, the orientation of the face, the number of faces, etc., indicates the frequency of a specific gender estimated from the face, the frequency of a specific age estimated from the face, or the degree) determines the selection set.

在本发明的第六方式构成为包括上述分类单元和集合决定单元的情况下，决定单元构成为根据选择集合内所包含的人脸，从选择集合内所包含的人脸中决定成为对象的脸。When the sixth aspect of the present invention is configured to include the classification unit and the set determination unit, the determination unit is configured to determine a face to be a target from the faces included in the selection set based on the faces included in the selection set. .

根据这样构成的本发明的第六方式，图像中的多个脸被分为多个集合，从多个集合中决定用于决定成为处理对象的脸的选择集合。然后，决定单元从所决定的选择集合内所包含的脸中最终决定成为处理对象的脸。According to the sixth aspect of the present invention thus constituted, a plurality of faces in an image are divided into a plurality of sets, and a selection set for determining a face to be processed is determined from the plurality of sets. Then, the determination unit finally determines a face to be processed from among the faces included in the determined selection set.

因此，在图像中存在多个脸的集合的情况下，不是根据多个集合内所包含的所有脸来决定成为处理对象的脸，而是仅根据其中任意一个集合内所包含的脸来决定成为处理对象的脸。因此，可进行仅与由集合决定单元所决定的选择集合内所包含的脸对应的控制。例如，在图像中有多个人的集体相分离而存在的情况下，可对任意一个集体进行特定的对焦和/或曝光控制，而不是对这些集体进行平均的处理。并且，例如在图像中存在集体和孤立的人的情况下，可进行不包含该孤立的人而仅特别针对集体，或者仅特别针对孤立的人的对焦和/或曝光控制。Therefore, when there are multiple sets of faces in the image, the face to be processed is not determined based on all the faces contained in the multiple sets, but the face to be processed is determined only based on the faces contained in any one of the sets. Process the subject's face. Therefore, it is possible to perform control corresponding to only the faces included in the selection set determined by the set determination unit. For example, in the case where multiple groups of people exist separately in an image, specific focus and/or exposure control may be performed on any one group instead of being averaged over these groups. Also, for example, when there are collective and isolated persons in the image, focus and/or exposure control may be performed specifically for only the collective or only for the isolated persons without including the isolated persons.

本发明的第六方式可以构成为还包括显示单元，该显示单元把由决定单元所决定的人脸与其它脸区别开来显示。显示单元通过例如在所决定的人脸的周围显示框，与其它脸区别开来显示。另外，显示单元通过例如在所决定的人脸的周围显示与在其它脸的周围所显示的框具有不同的颜色、粗细或形状的框，与其它脸区别开来显示。另外，显示单元例如通过对所决定的人脸实施与其它脸不同的图像处理并显示，与其它脸区别开来显示。A sixth aspect of the present invention may be configured to further include display means for displaying the human face determined by the determination means in distinction from other faces. The display unit displays, for example, a frame around the determined face to distinguish it from other faces. In addition, the display unit displays, for example, a frame having a different color, thickness, or shape around the determined face from those displayed around the other faces to distinguish it from other faces. In addition, the display unit displays the determined face differently from other faces by subjecting it to image processing different from other faces, for example.

〔第七方式〕[Seventh method]

本发明的第七方式是一种对象决定装置，包括检测单元和决定单元。其中，检测单元采用与本发明的第六方式相同的结构。A seventh aspect of the present invention is an object determination device including detection means and determination means. Among them, the detection unit has the same configuration as that of the sixth aspect of the present invention.

本发明的第七方式中的决定单元在由检测单元检测出多个人脸时，以所检测的脸的数目为基准把位于其正中的人脸决定为成为对焦或曝光控制的对象的脸。该决定单元对所检测出的脸的数目进行计数，根据所计数的数的中间值来决定脸。例如，该决定单元将各脸根据其x坐标进行排序，把上述与中间值的顺序相当的脸决定为成为对象的脸。该决定单元也可以根据y坐标而不是x坐标来决定成为对象的脸。并且，该决定单元可以构成为根据x坐标和y坐标分别选择脸，在选择了同一个脸的情况下，把该脸决定为成为对象的脸，在选择了不同的脸的情况下，根据规定的基准把其中任意一个脸决定为成为对象的脸。In the seventh aspect of the present invention, when a plurality of human faces are detected by the detection unit, the determination unit determines the central human face as the face to be controlled by focusing or exposure based on the number of detected faces. The determining unit counts the number of detected faces, and determines a face based on an intermediate value of the counted numbers. For example, the determination unit sorts the faces according to their x-coordinates, and determines the face corresponding to the order of the median value as the target face. The determination unit may determine the face to be the subject based on the y coordinate instead of the x coordinate. In addition, the determination unit may be configured to select a face according to the x coordinate and the y coordinate, and if the same face is selected, the face is determined as the target face; Any one of the faces is determined as the target face based on the reference.

根据本发明的第七方式，可类似地取得与构成为根据多个脸的位置的中心来决定成为对象的脸的本发明的第六方式相同的效果。并且，根据本发明的第七方式，没有必要取得多个脸的位置的中心。即，只要执行脸的计数和数据排序就够了，没有必要执行几何计算。因此，在使用通用的信息处理装置的情况下，与本发明的第六方式相比可快速地执行处理。According to the seventh aspect of the present invention, the same effect as that of the sixth aspect of the present invention configured to determine a target face based on the centers of positions of a plurality of faces can be similarly obtained. Furthermore, according to the seventh aspect of the present invention, it is not necessary to obtain the centers of the positions of a plurality of faces. That is, it is sufficient to perform face counting and data sorting, and it is not necessary to perform geometry calculations. Therefore, when a general-purpose information processing device is used, processing can be performed more quickly than in the sixth aspect of the present invention.

并且，本发明的第七方式可以与本发明的第六方式一样，构成为还包括分类单元和集合决定单元。在该情况下，本发明的第七方式中的决定单元根据选择集合内所包含的人脸，决定成为对象的脸。Furthermore, the seventh aspect of the present invention may be configured to further include classification means and aggregation determination means, as in the sixth aspect of the present invention. In this case, the determination unit in the seventh aspect of the present invention determines a target face based on human faces included in the selection set.

〔第八方式〕[Eighth way]

本发明的第八方式是一种对象决定装置，包括检测单元、分类单元、暂定决定单元以及最终决定单元。其中，检测单元和分类单元采用与本发明的第六方式相同的结构。An eighth aspect of the present invention is an object determination device including a detection unit, a classification unit, a provisional determination unit, and a final determination unit. Among them, the detection unit and the classification unit have the same configuration as the sixth aspect of the present invention.

本发明的第八方式中的暂定决定单元在由分类单元所生成的多个集合中的各个集合中，暂定决定成为对焦或曝光控制的对象的脸。换句话说，暂定决定单元针对多个集合中的各个集合，从该集合内所包含的人脸中，决定在实施摄像时成为进行对焦的对象的脸和/或成为进行曝光控制的对象的脸。此时，暂定决定单元根据哪种基准来决定成为对象的脸均可。例如，暂定决定单元可以通过与本发明的第六方式或第七方式所示的决定单元相同的处理，来决定成为对象的脸。另外，例如，暂定决定单元可以根据本发明的第六方式的脸决定单元的规定基准，来决定成为对象的脸。The provisional determination unit in the eighth aspect of the present invention provisionally determines a face to be a focus or exposure control target in each of the plurality of sets generated by the classification unit. In other words, for each of the plurality of sets, the provisional determination unit determines, from the faces included in the set, a face to be focused on when imaging is performed and/or a face to be subjected to exposure control. Face. At this time, the provisional determination unit may determine the target face based on any criterion. For example, the tentative determining means may determine the target face by the same process as the determining means described in the sixth or seventh aspect of the present invention. In addition, for example, the provisional determination unit may determine the target face based on the predetermined criteria of the face determination unit according to the sixth aspect of the present invention.

本发明的第八方式中的最终决定单元从由暂定决定单元所决定的脸中最终决定成为对焦或曝光控制的对象的脸。例如，最终决定单元可以通过与本发明的第六方式或第七方式所示的决定单元相同的处理，从各集合中由暂定决定单元所决定的脸中最终决定成为对象的脸。另外，例如，最终决定单元可以根据本发明的第六方式的脸决定单元的规定基准，从各集合中由暂定决定单元所决定的脸中，最终决定成为对象的脸。The final determination unit in the eighth aspect of the present invention finally determines a face to be a focus or exposure control target from the faces determined by the tentative determination unit. For example, the final determination unit may finally determine the target face from the faces determined by the tentative determination unit in each set by the same process as the determination unit described in the sixth or seventh aspect of the present invention. Also, for example, the final determination unit may finally determine the target face from the faces determined by the provisional determination unit in each set based on the predetermined criteria of the face determination unit according to the sixth aspect of the present invention.

根据本发明的第八方式，可获得与构成为还包括分类单元和集合决定单元的本发明的第六方式或第七方式相同的效果。并且，根据本发明的第八方式，在维持因包括分类单元和集合决定单元而获得的效果的状态下，可以在决定成为对象的脸时根据各种基准来决定该脸。According to the eighth aspect of the present invention, the same effects as those of the sixth or seventh aspect of the present invention configured to further include the classifying means and the aggregation determining means can be obtained. Furthermore, according to the eighth aspect of the present invention, it is possible to determine a target face based on various criteria while maintaining the effect obtained by including the classification means and the aggregation determination means.

〔第九方式〕[Ninth way]

本发明的第九方式是一种对象决定装置，包括检测单元、区决定单元以及决定单元。A ninth aspect of the present invention is an object determination device including detection means, area determination means, and determination means.

本发明的第九方式中的检测单元在将所输入的图像进行分割而得到的多个区的各区中检测人脸。可以预先决定将所输入的图像分成什么样的区，也可以动态决定。The detection unit in the ninth aspect of the present invention detects a human face in each of a plurality of areas obtained by dividing an input image. What kind of regions to divide the input image into can be determined in advance, or can be determined dynamically.

区决定单元根据检测单元的检测结果，决定用于决定在实施摄像时成为进行对焦的对象的脸和/或成为进行曝光控制的对象的脸的区(即包含成为对焦或曝光控制的对象的脸的区：称为“选择区”)。例如，可以根据与本发明的第六方式中的集合决定单元相同的基准来决定区。The region determination unit determines a region (that is, a region including a face to be focused or an exposure control) for determining a face to be focused on and/or a face to be subjected to exposure control during imaging based on the detection result of the detection unit. area: called the "selection area"). For example, the region may be determined based on the same criteria as the aggregate determination means in the sixth aspect of the present invention.

本发明的第九方式中的决定单元从选择区内所包含的脸中决定成为对象的脸。此时，该决定单元可以根据哪种基准来决定成为对象的脸均可。例如，该决定单元可以通过与本发明的第六方式或第七方式所示的决定单元相同的处理，来决定成为对象的脸。另外，例如，该决定单元可以根据本发明的第六方式的脸决定单元的规定基准，决定成为对象的脸。The determining means in the ninth aspect of the present invention determines a target face from faces included in the selection area. At this time, the determining unit may determine the target face based on any reference. For example, the determining unit may determine the target face by the same process as the determining unit described in the sixth or seventh aspect of the present invention. Also, for example, the determination unit may determine the target face based on the predetermined criteria of the face determination unit according to the sixth aspect of the present invention.

〔第十方式〕[Tenth method]

本发明的第十方式是一种对象决定装置，包括检测单元、判断单元、选择单元以及决定单元。其中，检测单元采用与本发明的第六方式相同的结构。A tenth aspect of the present invention is an object determination device including detection means, determination means, selection means, and determination means. Among them, the detection unit has the same configuration as that of the sixth aspect of the present invention.

判断单元在检测单元检测出多个脸时，判断所检测出的人脸中的最大的脸。判断单元可以根据例如所检测出的人脸中的肤色的像素数来判断大小。另外，判断单元可以根据例如在被检测为人脸时所使用的脸矩形的大小来判断脸的大小。判断单元构成为根据其它哪种基准来判断脸的大小均可。另外，最大的脸没有必要是严格意义上的最大，可以是例如第2大的脸，也可以是第3大的脸。The judging unit judges the largest face among the detected human faces when the detecting unit detects a plurality of faces. The judging unit may judge the size based on, for example, the number of pixels of the detected skin color in the human face. In addition, the judging unit may judge the size of the face based on, for example, the size of a face rectangle used when being detected as a human face. The judging unit may be configured to judge the size of the face based on any other criteria. In addition, the largest face does not have to be the largest in the strict sense, and may be, for example, the second largest face or the third largest face.

选择单元选择所检测出的脸中的最大的脸、以及具有基于该脸的大小的规定范围内的大小的其它脸。The selection unit selects the largest face among the detected faces and other faces having a size within a predetermined range based on the size of the face.

本发明的第十方式中的决定单元从由选择单元所选择的脸中，决定在实施摄像时成为进行对焦的对象的脸和/或成为进行曝光控制的对象的脸。这样，本发明的第十方式中的决定单元不是根据由检测单元所检测出的多个脸全部的位置来决定成为对象的脸，而是从所检测出的脸中的最大的脸和具有基于该脸的大小的规定范围内的大小的其它脸中，决定成为对象的脸。该决定单元根据哪种基准决定成为对象的脸均可。例如，该决定单元可以通过与本发明的第六方式或第七方式所示的决定单元相同的处理，来决定成为对象的脸。另外，例如，该决定单元可以根据本发明的第六方式的脸决定单元的规定基准，决定成为对象的脸。The determination unit in the tenth aspect of the present invention determines a face to be focused on and/or a face to be controlled for exposure from the faces selected by the selection unit. In this way, the determination unit in the tenth aspect of the present invention does not determine the target face based on the positions of all the faces detected by the detection unit, but selects the largest face among the detected faces A target face is determined among other faces having a size within a predetermined range of the size of the face. The determining unit may determine the target face based on any criteria. For example, the determining unit may determine the target face by the same process as the determining unit described in the sixth or seventh aspect of the present invention. Also, for example, the determination unit may determine the target face based on the predetermined criteria of the face determination unit according to the sixth aspect of the present invention.

根据本发明的第十方式，仅根据在对焦用图像内具有相对某种程度的大小的脸，选择成为对焦对象的脸。因此，可防止把被拍摄为背景的人，即用户没有有意作为被摄物体的人脸作为处理对象而包含进去。因此，例如在该决定单元进行与本发明的第六方式中的决定单元相同的处理的情况下，提高了选择包含成为对焦对象的脸的集合的处理、取得基于集合内所包含的脸的中心的处理、以及选择成为对焦对象的脸的处理的精度。According to the tenth aspect of the present invention, the face to be focused is selected based on only faces having a relatively certain size in the focusing image. Therefore, it is possible to prevent a person photographed as the background, that is, a face that the user does not intend to be a subject, from being included as a processing target. Therefore, for example, when the determination unit performs the same processing as that of the determination unit in the sixth aspect of the present invention, the process of selecting a set including a face to be focused on and obtaining a center based on the faces included in the set are improved. processing, and the precision of processing for selecting the face to be in focus.

〔第十一方式〕[Eleventh way]

本发明的第十一方式是一种对象决定装置，包括检测单元、分类单元、集合决定单元以及决定单元。其中，检测单元、分类单元以及集合决定单元采用与第六方式相同的结构。An eleventh aspect of the present invention is an object determination device including a detection unit, a classification unit, an aggregate determination unit, and a determination unit. Among them, the detecting unit, the classifying unit, and the set determining unit adopt the same structure as that of the sixth embodiment.

本发明的第十一方式中的决定单元从由集合决定单元所决定的集合内所包含的脸中决定成为对象的脸。此时，该决定单元根据哪种基准决定成为对象的脸均可。例如，该决定单元可以通过与本发明的第六方式或第七方式所示的决定单元相同的处理，决定成为对象的脸。另外，例如，该决定单元可以根据本发明的第六方式的脸决定单元的规定基准，决定成为对象的脸。The determination unit in the eleventh aspect of the present invention determines a target face from faces included in the set determined by the set determination unit. In this case, the determining unit may determine the target face based on any criterion. For example, the determining unit may determine the target face by the same process as the determining unit described in the sixth or seventh aspect of the present invention. Also, for example, the determination unit may determine the target face based on the predetermined criteria of the face determination unit according to the sixth aspect of the present invention.

根据本发明的第十一方式，可取得与本发明的第八方式相同的效果。According to the eleventh aspect of the present invention, the same effect as that of the eighth aspect of the present invention can be obtained.

〔第十二方式〕[Twelfth method]

本发明的第十二方式是一种对象决定装置，包括检测单元、特征量取得单元以及决定单元。其中，检测单元采用与本发明的第六方式相同的结构。A twelfth aspect of the present invention is an object determination device including detection means, feature quantity acquisition means, and determination means. Among them, the detection unit has the same configuration as that of the sixth aspect of the present invention.

特征量取得单元针对在检测单元中所检测出的多个人脸的各个，取得表示人脸程度的特征量。“入脸程度”可以根据例如距用于划分某图像是否是人脸的识别边界线的距离来给出。人脸程度可以使用在检测单元中的脸检测处理时所取得的值来表示。The feature quantity acquisition unit acquires a feature quantity indicating a degree of a face for each of the plurality of faces detected by the detection unit. The "face-introduction degree" can be given according to, for example, the distance from the recognition boundary line used to classify whether an image is a human face or not. The degree of a human face can be expressed using a value obtained during face detection processing in the detection unit.

决定单元根据特征量，决定在实施摄像时成为进行对焦的对象的脸和/或成为进行曝光控制的对象的脸。The determining unit determines a face to be focused on and/or a face to be controlled for exposure based on the feature amount.

〔第六方式～第十二方式的变形〕[Variations of the sixth form to the twelfth form]

本发明的第六～十二方式可以构成为还包括帧取得单元，其从所输入的包含多个帧的时序图像中取得某一帧的图像作为处理对象的图像。在这样构成的情况下，检测单元从由帧取得单元所取得的某一帧的图像中检测人脸。The sixth to twelfth aspects of the present invention may be configured to further include a frame acquisition unit that acquires an image of a certain frame from the input time-series images including a plurality of frames as an image to be processed. In such a configuration, the detection unit detects a human face from an image of a certain frame acquired by the frame acquisition unit.

在这样构成的本发明的第六～十二方式中，对于包含多个帧的时序图像，可获得与本发明的第六～十二方式相同的作用/效果。In the sixth to twelfth aspects of the present invention constituted in this way, the same operations/effects as those of the sixth to twelfth aspects of the present invention can be obtained for time-series images including a plurality of frames.

〔其它〕〔other〕

本发明的第一方式～第十二方式可以通过使用信息处理装置执行程序来实现。即，本发明可把上述第一方式～第十二方式中的各单元所执行的处理确定为用于使信息处理装置执行的程序、或者记录有该程序的记录介质。The first to twelfth aspects of the present invention can be realized by executing a program using an information processing device. That is, in the present invention, the processing performed by each means in the above-mentioned first to twelfth aspects can be specified as a program for the information processing device to execute, or a recording medium on which the program is recorded.

根据本发明，当检测被摄物体时，作为预备摄像没有必要摄像一定对该被摄物体进行了对焦的图像。因此，在预备摄像中不进行用于取得多个进行了对焦的图像等的焦点控制，仅使用根据规定的焦点信息所取得的图像就能快速地检测主要被摄物体。According to the present invention, when detecting a subject, it is not necessary to capture an image focusing on the subject as a preliminary imaging. Therefore, the main subject can be quickly detected using only images acquired based on predetermined focus information without performing focus control for acquiring a plurality of focused images during preliminary imaging.

并且，根据本发明，在预备摄像中检测出多个人物的脸的情况下，根据所检测出的多个脸的位置和大小等，自动判断应对哪个脸进行对焦或曝光控制。Furthermore, according to the present invention, when a plurality of human faces are detected during preliminary imaging, it is automatically determined which face should be focused or exposed based on the positions, sizes, etc. of the detected faces.

附图说明Description of drawings

图1是表示摄像装置的功能块的图。FIG. 1 is a diagram showing functional blocks of an imaging device.

图2是表示使用脸判定矩形的脸的检测例的图。FIG. 2 is a diagram showing an example of face detection using a face determination rectangle.

图3是表示脸判定矩形的图形例的图。FIG. 3 is a diagram showing a graphic example of a face determination rectangle.

图4是表示使用脸判定矩形的脸的检测例的图。FIG. 4 is a diagram showing an example of face detection using a face determination rectangle.

图5是表示摄像装置的动作例的流程图。FIG. 5 is a flowchart showing an example of the operation of the imaging device.

图6是表示对焦对象决定部的动作例的流程图。FIG. 6 is a flowchart showing an example of an operation of a focus target determination unit.

图7是表示摄像装置的处理例的图。FIG. 7 is a diagram showing an example of processing by the imaging device.

图8是表示第一实施方式的功能块的图。FIG. 8 is a diagram showing functional blocks of the first embodiment.

图9是用于对外接多边形法进行说明的图。FIG. 9 is a diagram for explaining the circumscribed polygon method.

图10是用于对重心法进行说明的图。FIG. 10 is a diagram for explaining the center-of-gravity method.

图11是用于对脸决定部的处理进行说明的图。FIG. 11 is a diagram for explaining processing of a face determining unit.

图12是表示第二实施方式的动作例的流程图。FIG. 12 is a flowchart showing an example of operation of the second embodiment.

图13是表示第二实施方式的动作例的一部分的流程图。FIG. 13 is a flowchart showing a part of an operation example of the second embodiment.

图14是表示第二实施方式的动作例的一部分的流程图。FIG. 14 is a flowchart showing a part of an operation example of the second embodiment.

图15是表示第三实施方式的功能块的图。FIG. 15 is a diagram showing functional blocks of the third embodiment.

图16是表示分类部的处理例的图。FIG. 16 is a diagram showing an example of processing by a classification unit.

图17是表示第三实施方式中的决定部的处理例的图。FIG. 17 is a diagram showing an example of processing by a determination unit in the third embodiment.

图18是表示第三实施方式的动作例的一部分的流程图。FIG. 18 is a flowchart showing a part of an operation example of the third embodiment.

图19是用于表示第三实施方式中的效果的图。FIG. 19 is a diagram showing effects in the third embodiment.

图20是表示第四实施方式的功能块的图。FIG. 20 is a diagram showing functional blocks of the fourth embodiment.

图21是表示第四实施方式的动作例的一部分的流程图。FIG. 21 is a flowchart showing a part of an operation example of the fourth embodiment.

图22是表示第五实施方式的功能块的图。FIG. 22 is a diagram showing functional blocks of the fifth embodiment.

图23是表示被分为多个区的对焦用图像示例的图。FIG. 23 is a diagram showing an example of an in-focus image divided into a plurality of areas.

图24是表示第五实施方式中的决定部的处理例的图。FIG. 24 is a diagram showing an example of processing by a determination unit in the fifth embodiment.

图25是表示第五实施方式的动作例的一部分的流程图。FIG. 25 is a flowchart showing a part of an operation example of the fifth embodiment.

图26是表示第六实施方式的功能块的图。FIG. 26 is a diagram showing functional blocks of the sixth embodiment.

图27是表示最大脸选择部的处理例的图。FIG. 27 is a diagram showing an example of processing by a maximum face selection unit.

图28是表示候选选择部的处理例的图。FIG. 28 is a diagram showing an example of processing by a candidate selection unit.

图29是表示第六实施方式的动作例的一部分的流程图。FIG. 29 is a flowchart showing a part of an operation example of the sixth embodiment.

符号说明Symbol Description

1：摄像装置；2：摄像部；3：图像存储部；4：对焦对象决定部；5：脸检测部；6：决定部；7：测距部；8：脸判定矩形；9：第一矩形；10：第二矩形；11：输入部；12a、12b、12c、12d、12e：摄像装置；13：输入部；14：摄像部；15：图像存储部；16a、16b、16c、16d、16e：对焦对象决定部；17a、17d：脸检测部；18a、18b、18c、18d、18e：决定部；19a、19b、19d、19e：中心决定部；20a、20b、20d、20e：脸决定部；21：测距部；22：分类部；23：集合决定部；24：区决定部；25：最大脸选择部；26：候选选择部。1: camera device; 2: camera unit; 3: image storage unit; 4: focus target determination unit; 5: face detection unit; 6: determination unit; 7: distance measurement unit; 8: face determination rectangle; 9: first 10: second rectangle; 11: input unit; 12a, 12b, 12c, 12d, 12e: camera; 13: input unit; 14: camera unit; 15: image storage unit; 16e: focus target determination unit; 17a, 17d: face detection unit; 18a, 18b, 18c, 18d, 18e: determination unit; 19a, 19b, 19d, 19e: center determination unit; 20a, 20b, 20d, 20e: face determination 21: ranging section; 22: classification section; 23: set decision section; 24: area decision section; 25: maximum face selection section; 26: candidate selection section.

具体实施方式Detailed ways

[第一实施方式][first embodiment]

下面，使用图对第一实施方式的摄像装置1进行说明。另外，针对摄像装置1的以下说明仅是例示而已，其结构不限于以下的说明。Next, theimaging device 1 of the first embodiment will be described with reference to the drawings. In addition, the following description of theimaging device 1 is merely an example, and its configuration is not limited to the following description.

〔系统结构〕〔system structure〕

摄像装置1是数字相机、数字摄像机、或者具有数字相机和数字摄像机那样的摄像功能的信息处理装置。摄像装置1在硬件上具有通过总线连接的CPU(中央运算处理装置)、主存储装置(RAM(Read Only Memory，只读存储器))、辅助存储装置以及用于作为数字相机或数字摄像机而动作的各装置(摄像镜头、机械机构、CCD(Charge-Coupled Device，电荷耦合装置)、操作部、马达等)等。辅助存储装置使用非易失性存储装置而构成。这里所说的非易失性存储装置是指所谓的ROM(Read-OnlyMemory，只读存储器：包含EPROM(Erasable Programmable Read-OnlyMemory，可擦除可编程只读存储器)、EEPROM(Electrically ErasableProgrammable Read-Only Memory，电可擦除可编程只读存储器)、掩模型ROM等)、FRAM(Ferroelectric RAM，铁电RAM)、硬盘等。Theimaging device 1 is a digital camera, a digital video camera, or an information processing device having an imaging function such as a digital camera or a digital video camera. Theimaging device 1 has a CPU (central processing unit) connected via a bus, a main storage device (RAM (Read Only Memory, read-only memory)), an auxiliary storage device, and a device for operating as a digital camera or a digital video camera. Various devices (camera lens, mechanical mechanism, CCD (Charge-Coupled Device, charge-coupled device), operation unit, motor, etc.), etc. The auxiliary storage device is configured using a nonvolatile storage device. The non-volatile storage device mentioned here refers to the so-called ROM (Read-Only Memory, read-only memory: including EPROM (Erasable Programmable Read-Only Memory, erasable programmable read-only memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory) Memory, electrically erasable programmable read-only memory), mask ROM, etc.), FRAM (Ferroelectric RAM, ferroelectric RAM), hard disk, etc.

图1是表示摄像装置1的功能块的图。摄像装置1通过把存储在辅助存储装置内的各种程序(OS、应用程序等)加载到主存储装置内并通过CPU来执行，从而作为包括摄像部2、图像存储部3、对焦对象决定部4、测距部7以及输入部11等的装置来发挥功能。对焦对象决定部4是通过使用CPU来执行对焦对象决定程序而实现的。另外，对焦对象决定部4也可以作为专用硬件来实现。FIG. 1 is a diagram showing functional blocks of theimaging device 1 . Theimaging device 1 loads various programs (OS, application programs, etc.) stored in the auxiliary storage device into the main storage device and executes them through the CPU, thereby comprising animaging unit 2, animage storage unit 3, and a focus target determination unit. 4. Devices such as thedistance measuring unit 7 and theinput unit 11 function. The focustarget determination unit 4 is realized by executing a focus target determination program using a CPU. In addition, the focustarget determination unit 4 may also be realized as dedicated hardware.

以下，使用图1对摄像装置1所包括的各功能部进行说明。Hereinafter, each functional unit included in theimaging device 1 will be described using FIG. 1 .

〔〔摄像部〕〕[[camera department]]

摄像部2使用摄像镜头、机械机构、CCD、马达等构成为具有自动聚焦功能的装置。摄像镜头包括例如用于实现变焦功能的变焦镜头、用于对任意被摄物体进行对焦的聚焦镜头等。机械机构包括机械快门、光圈、滤光器等。马达包括变焦镜头马达、聚焦马达、快门马达等。Theimaging unit 2 is configured as a device having an autofocus function using an imaging lens, a mechanism, a CCD, a motor, and the like. The imaging lens includes, for example, a zoom lens for realizing a zoom function, a focus lens for focusing on an arbitrary subject, and the like. Mechanical mechanisms include mechanical shutters, apertures, filters, etc. The motor includes a zoom lens motor, a focus motor, a shutter motor, and the like.

摄像部2具有未作图示的存储部，把规定的焦点信息存储在该存储部内。规定的焦点信息是预先决定的一个焦点信息。在摄像装置1的电源接通的情况下，如果从测距部7没有焦点信息的输入，则摄像部2的聚焦镜头被控制成成为基于该规定的焦点信息的状态。另一方面，如果从测距部7有焦点信息的输入，则摄像部2的聚焦镜头被控制成成为基于所输入的焦点信息的状态。Theimaging unit 2 has a storage unit (not shown), and stores predetermined focus information in the storage unit. The prescribed focus information is a piece of focus information determined in advance. When theimaging device 1 is powered on, if there is no input of focus information from thedistance measuring unit 7 , the focus lens of theimaging unit 2 is controlled to be in a state based on the predetermined focus information. On the other hand, when focus information is input from thedistance measuring unit 7, the focus lens of theimaging unit 2 is controlled to be in a state based on the input focus information.

摄像部2使用CCD把通过包括像上述那样被控制的聚焦镜头的摄像镜头而成像的被摄物体的图像转换成电信号，从而进行摄像。Theimaging unit 2 converts an image of a subject formed by an imaging lens including a focus lens controlled as described above into an electrical signal using a CCD, and performs imaging.

摄像部2可以构成为每当用户通过输入部11进行了输入时，就摄像1张或多张静态图像。并且，摄像部2也可以构成为根据用户通过输入部11进行的输入，摄像由多个帧构成的动态图像(时序图像)。并且，摄像部2在构成为摄像动态图像的情况下，可以构成为从多个帧中取得一个帧的图像，把该帧的图像作为处理对象的图像传递给对焦对象决定部4。Theimaging unit 2 may be configured to capture one or more still images every time the user inputs through theinput unit 11 . Furthermore, theimaging unit 2 may be configured to capture a moving image (time-series image) composed of a plurality of frames in accordance with an input by the user through theinput unit 11 . Furthermore, when theimaging unit 2 is configured to capture a moving image, it may be configured to obtain an image of one frame from a plurality of frames, and pass the image of this frame to the focustarget determination unit 4 as an image to be processed.

〔〔图像存储部〕〕[[Image storage unit]]

图像存储部3使用所谓的RAM等可读写记录介质等来构成。图像存储部3也可以使用相对于摄像装置1可拆装的记录介质来构成。图像存储部3存储由摄像部2所摄像的图像的数据。Theimage storage unit 3 is configured using a readable and writable recording medium such as a so-called RAM. Theimage storage unit 3 may also be configured using a recording medium that is detachable from theimaging device 1 . Theimage storage unit 3 stores data of images captured by theimaging unit 2 .

〔〔对焦对象决定部〕〕[[Focus Object Determination Unit]]

对焦对象决定部4从由摄像部2所摄像的被摄物体中决定进行对焦的被摄物体，即成为对焦的对象的被摄物体。此时，对焦对象决定部4使用根据规定的焦点信息由摄像部2为了对焦用而摄像的图像来进行处理。该为了对焦用而摄像的图像由摄像部2输入到对焦对象决定部4中。The focustarget determination unit 4 determines a subject to be focused on, that is, a subject to be focused on, from among subjects captured by theimaging unit 2 . At this time, the focustarget determination unit 4 performs processing using an image captured by theimaging unit 2 for focusing based on predetermined focus information. The image captured for focusing is input to the focustarget determination unit 4 by theimaging unit 2 .

对焦对象决定部4包括：脸检测部5，其从所输入的图像中检测成为被摄物体的人物的脸；以及决定部6，其根据由脸检测部5所检测出的脸决定进行对焦的被摄物体。以下，对构成对焦对象决定部4的各功能部进行说明。The focustarget determination part 4 includes: aface detection part 5, which detects the face of a person who becomes a subject from an input image; subject. Hereinafter, each functional unit constituting the in-focustarget determination unit 4 will be described.

〈脸检测部〉〈Face detection part〉

脸检测部5根据多个部分区域的统计量(具体地说是浓淡值)，从所输入的图像中检测成为被摄物体的人物的脸。以下，对由脸检测部5所执行的、检测人物的脸时的处理进行详细说明。Theface detection unit 5 detects the face of a person to be photographed from an input image based on statistics (specifically, shade values) of a plurality of partial regions. Hereinafter, the processing performed by theface detection unit 5 to detect a person's face will be described in detail.

图2是表示基于浓淡值的人物的脸的检测方法的图。在该检测方法中，使特定矩形(以下称为“脸判定矩形”)8在成为处理对象的图像内移动，判断在各移动目的地的脸判定矩形8内(以下称为“关注区域”)是否包含有人物的脸。此时，通过改变脸判定矩形8的大小和成为处理对象的图像的大小的比率、脸判定矩形8和成为处理对象的图像的角度，可检测各种大小或角度的脸。FIG. 2 is a diagram illustrating a method of detecting a person's face based on gradation values. In this detection method, a specific rectangle (hereinafter referred to as a "face determination rectangle") 8 is moved within an image to be processed, and it is judged that the area within the face determination rectangle 8 (hereinafter referred to as a "region of interest") at each moving destination is Whether to include a character's face. At this time, by changing the ratio between the size of theface determination rectangle 8 and the size of the image to be processed, and the angle between theface determination rectangle 8 and the image to be processed, faces of various sizes and angles can be detected.

在脸判定矩形8的矩形内包含多个其它矩形(以下称为“第一矩形”、“第二矩形”)9、10。第一矩形9和第二矩形10分别被配置在脸判定矩形8内的固定位置。在一个脸判定矩形8内分别配置1个以上第一矩形9和第二矩形10。A plurality of other rectangles (hereinafter referred to as “first rectangle” and “second rectangle”) 9 and 10 are contained within the rectangle of theface determination rectangle 8 . Thefirst rectangle 9 and thesecond rectangle 10 are respectively arranged at fixed positions within theface determination rectangle 8 . One or morefirst rectangles 9 andsecond rectangles 10 are respectively arranged within oneface determination rectangle 8 .

在脸的检测处理中，算出在各关注区域中，被第一矩形9和第二矩形10所包围的各个区域(以下分别称为“第一特征区域”、“第二特征区域”)的浓淡值。第一特征区域和第二特征区域的浓淡差例如由各区域内的像素值的平均值或合计值来表示。In the face detection process, the shades of the regions surrounded by thefirst rectangle 9 and the second rectangle 10 (hereinafter respectively referred to as "first characteristic region" and "second characteristic region") in each region of interest are calculated. value. The difference in gradation between the first characteristic region and the second characteristic region is represented by, for example, an average value or a total value of pixel values in each region.

然后，算出第一特征区域的浓淡值La和第二特征区域的浓淡值Lb的差量。然后，根据该差量值是否大于预先设定的阈值α，来判定在关注区域内是否包含有人物的脸。Then, the difference between the gradation value La of the first characteristic region and the gradation value Lb of the second characteristic region is calculated. Then, according to whether the difference value is greater than a preset threshold α, it is determined whether the face of the person is included in the attention region.

图3是表示脸判定矩形8的图形例的图。图4是表示使用多个图形的脸判定矩形8的脸判定处理的图。在实际处理中，如图3所示，为脸判定矩形8准备有多个图形。在各图形中，第一矩形9和第二矩形10的数量和位置各不相同。FIG. 3 is a diagram showing an example of a pattern of theface determination rectangle 8 . FIG. 4 is a diagram showing face determination processing using aface determination rectangle 8 of a plurality of graphics. In actual processing, as shown in FIG. 3 , a plurality of figures are prepared for theface determination rectangle 8 . In each figure, the number and position of thefirst rectangle 9 and thesecond rectangle 10 are different.

然后，针对各图形的脸判定矩形8通过计算浓淡差来执行判定。并且，根据各判定结果，针对每个关注区域算出得分，根据所算出的得分，对该关注区域内是否包含有人物的脸进行最终的判定。所算出的得分与关注区域内是否包含有人物的脸的对应是通过例如使用数量较多的样本图像(包含人物的脸的图像和不包含人物的脸的图像)的学习处理而预先获得的。Then, determination is performed by calculating the difference in gradation for theface determination rectangle 8 of each figure. Then, based on the determination results, a score is calculated for each ROI, and based on the calculated score, a final determination is made as to whether or not the face of the person is included in the ROI. The correspondence between the calculated score and whether a person's face is included in the region of interest is obtained in advance by, for example, learning processing using a large number of sample images (images including a person's face and images not including a person's face).

脸检测部5在检测出脸的情况下，即进行了在某关注区域内包含有人物的脸的最终判定的情况下，把与该脸有关的图像信息输出到决定部6。具体地说，脸检测部5输出包含有检测出的脸的关注区域的位置和大小。例如，脸检测部5输出该关注区域的中心坐标和面积(区域内的像素数)。When theface detection unit 5 detects a face, that is, when it is finally determined that a face of a person is included in a certain region of interest, theface detection unit 5 outputs image information related to the face to thedetermination unit 6 . Specifically, theface detection unit 5 outputs the position and size of the region of interest including the detected face. For example, theface detection unit 5 outputs the center coordinates and area (the number of pixels in the region) of the region of interest.

并且，脸检测部5可以构成为根据需要对所检测出的脸的朝向进行判定并输出。另外，脸检测部5在判定脸的朝向的情况下，通过使用例如非专利文献1、2所记载的技术来进行判定。具体地说，脸检测部5还使用例如左朝向用的图形的脸判定矩形8和右朝向用的图形的脸判定矩形8来执行脸检测。并且，脸检测部5在使用例如左朝向用的图形的脸判定矩形8检测出脸的情况下，判定为所检测出的脸是左朝向。同样，脸检测部5在使用右朝向用的图形的脸判定矩形8检测出脸的情况下，判定为所检测出的脸是右朝向。对于正面也一样。脸的朝向判定可以应用任何技术来实施而不限于上述技术。Furthermore, theface detection unit 5 may be configured to determine and output the direction of the detected face as necessary. In addition, when theface detection unit 5 determines the direction of the face, it determines by using, for example, the techniques described inNon-Patent Documents 1 and 2. Specifically, theface detection unit 5 also performs face detection using, for example, theface determination rectangle 8 of the figure for the left and theface determination rectangle 8 of the figure for the right. Furthermore, when a face is detected using, for example, theface determination rectangle 8 of a left-facing pattern, theface detection unit 5 determines that the detected face is facing left. Similarly, when a face is detected using theface determination rectangle 8 of the right-facing figure, theface detection unit 5 determines that the detected face is facing right. The same is true for the front. The determination of the orientation of the face can be implemented by applying any technique and is not limited to the techniques described above.

〈决定部〉〈decision department〉

决定部6根据从脸检测部5输入的与脸有关的图像信息，决定进行对焦的被摄物体。此时，大致来分，决定部6进行二个处理。第一处理是从由脸检测部5所检测出的多个脸中选择一个脸的处理。第二处理是根据由第一处理所选择的脸，决定进行对焦的被摄物体的处理。另外，在脸检测部5中仅检测出一个脸的情况下，省略第一处理。并且，在脸检测部5中未检测出脸的情况下，省略第一处理和第二处理。Thedetermination unit 6 determines the subject to focus on based on the image information on the face input from theface detection unit 5 . At this time, roughly divided, thedetermination unit 6 performs two processes. The first process is a process of selecting one face from a plurality of faces detected by theface detection unit 5 . The second process is a process of determining a subject to focus on based on the face selected by the first process. In addition, when only one face is detected by theface detection unit 5, the first processing is omitted. Also, when no face is detected by theface detection unit 5, the first processing and the second processing are omitted.

〈〈第一处理〉〉<<first processing>>

首先，对第一处理进行说明。在第一处理中，决定部6选择在第二处理中决定进行对焦的被摄物体时要使用哪个脸的图像信息。在第一处理中，有被称为面积比较法、得分比较法、测距法的三种实现方法。在决定部6中可以采用其中的任意一种方法。以下，对各实现方法进行说明。First, the first processing will be described. In the first process, thedetermination unit 6 selects which face image information is to be used when determining the subject to be focused on in the second process. In the first processing, there are three implementation methods called area comparison method, score comparison method, and distance measurement method. Any one of these methods can be adopted in thedecision unit 6 . Hereinafter, each implementation method will be described.

〈〈〈面积比较法〉〉〉<<<Area Comparison Method>>>

在面积比较法中，决定部6对在脸检测部5中所检测出的多个脸的面积进行比较，选择作为图像的面积为最大的脸。在面积比较法中，决定部6为了检测距摄像装置1最近的脸，选择作为图像的面积为最大的脸。即，在面积比较法中，决定部6通过对脸图像的大小进行比较来实施到被摄物体的距离的比较。In the area comparison method, thedetermination unit 6 compares the areas of the plurality of faces detected by theface detection unit 5 and selects the face having the largest area as an image. In the area comparison method, thedetermination unit 6 selects a face having the largest area as an image in order to detect the face closest to theimaging device 1 . That is, in the area comparison method, thedetermination unit 6 compares the sizes of the face images to compare the distances to the subjects.

〈〈〈得分比较法〉〉〉<<<Score Comparison Method>>>

在得分比较法中，决定部6针对在脸检测部5中所检测出的多个脸分别求出得分。决定部6根据脸的朝向、脸的大小(面积)、脸在图像内的位置、脸的周围的脸密度等获取得分。决定部6可以针对上述所有四项获得得分，也可以针对一部分项获取得分。In the score comparison method, thedetermination unit 6 obtains scores for each of the plurality of faces detected by theface detection unit 5 . Thedetermination unit 6 acquires a score based on the orientation of the face, the size (area) of the face, the position of the face in the image, the density of faces around the face, and the like. Thedetermination unit 6 may acquire points for all of the above four items, or may acquire points for some of the items.

然后，决定部6选择获得了最高得分的脸。关于脸的朝向，例如脸的朝向越是面向正面，则得分被设定得越高。关于脸的大小，例如脸的大小越大，则得分被设定得越高。关于脸在图像内的位置，例如距图像中央越近，则得分被设定得越高。关于脸密度，例如脸密度越高，则得分被设定得越高。然而，在各项(朝向、大小、位置、脸密度)中把哪种状态设定为高得分，可以由用户可变更地来设计，也可以由设计者任意设计。Then, thedetermination unit 6 selects the face that has obtained the highest score. Regarding the orientation of the face, for example, the more the orientation of the face faces the front, the higher the score is set. Regarding the size of the face, for example, the larger the size of the face, the higher the score is set. Regarding the position of the face within the image, for example, the closer the face is to the center of the image, the higher the score is set. Regarding the face density, for example, the higher the face density, the higher the score is set. However, which state to set as a high score in each item (orientation, size, position, and face density) may be variably designed by the user, or may be arbitrarily designed by the designer.

〈〈〈测距法〉〉〉〈〈Distance measurement method〉〉〉

意设计。intentional design.

〈〈〈测距法〉〉〉〈〈Distance measurement method〉〉〉

在测距法中，决定部6取得针对由脸检测部5所检测出的各脸的实物的测距结果。并且，决定部6选择距摄像装置1的距离最小的脸。In the distance measurement method, thedetermination unit 6 acquires the actual distance measurement result for each face detected by theface detection unit 5 . Then, thedetermination unit 6 selects the face whose distance from theimaging device 1 is the smallest.

决定部6可以构成为获得任意测距装置的测距结果。例如，可以构成为在摄像装置1或对焦对象决定部4中配备未作图示的测距装置，由决定部6获得该测距装置的测距结果。并且，例如，在测距部7采用对到被摄物体的距离进行测距的结构的情况下，可以把测距部7用作测距装置的替代装置。在该情况下，决定部6获得测距部7的测距结果(参照图1的虚线箭头)。在任一种情况下，决定部6把从脸检测部5输入的图像信息中的关注区域的坐标提供给测距装置。并且，测距装置根据该关注区域的坐标，对到所检测出的脸的距离进行测距。Thedetermination unit 6 may be configured to obtain a distance measurement result of an arbitrary distance measurement device. For example, a distance measuring device (not shown) may be provided in theimaging device 1 or the focustarget determination unit 4 , and thedetermination unit 6 may obtain the distance measurement result of the distance measuring device. Also, for example, in the case where thedistance measuring unit 7 is configured to measure the distance to the subject, thedistance measuring unit 7 can be used as a substitute for the distance measuring device. In this case, thedetermination unit 6 obtains the distance measurement result of the distance measurement unit 7 (see the dotted arrow in FIG. 1 ). In either case, thedetermination unit 6 supplies the coordinates of the region of interest in the image information input from theface detection unit 5 to the distance measuring device. Then, the distance measuring device measures the distance to the detected face based on the coordinates of the region of interest.

〈〈第二处理〉〉<<second processing>>

在第二处理中，决定部6根据所选择的脸，决定进行对焦的被摄物体。作为这种被摄物体的例子，有所选择的脸的部位(例如眼、鼻、口、耳、额等)。此时，所选择的脸的部位的位置可以由脸检测部5重新检测，也可以根据所选择的脸的图像信息求出其位置。例如，决定部6把在规定方向上从所选择的脸的中心坐标相对地相隔了规定距离的位置识别为所选择的脸的部位的位置。然后，决定部6把进行对焦的被摄物体的图像的坐标(即二维坐标)输出给测距部7。In the second process, thedetermination unit 6 determines a subject to be focused on based on the selected face. As an example of such a subject, selected face parts (for example, eyes, nose, mouth, ears, forehead, etc.). At this time, the position of the selected face part may be re-detected by theface detection unit 5, or the position may be obtained from the image information of the selected face. For example, thedetermination unit 6 recognizes a position relatively separated by a predetermined distance from the center coordinates of the selected face in a predetermined direction as the position of the selected face part. Then, thedetermination unit 6 outputs the coordinates (that is, the two-dimensional coordinates) of the image of the subject to be focused on to thedistance measuring unit 7 .

并且，所选择的脸本身可以是进行对焦的被摄物体。在该情况下，决定部6把在第一处理中所选择的脸的图像信息内所包含的脸的位置(即二维坐标)输出给测距部7。Also, the selected face itself may be the subject to be focused on. In this case, thedetermination unit 6 outputs the position (that is, the two-dimensional coordinates) of the face included in the image information of the face selected in the first process to thedistance measuring unit 7 .

〔〔测距部〕〕[[Distance Measuring Department]]

测距部7取得用于对由对焦对象决定部4的决定部6所决定的被摄物体对焦的焦点信息。此时，测距部7根据从决定部6输入的二维坐标，确定用于进行对焦的被摄物体。Thedistance measuring unit 7 acquires focus information for focusing on the subject determined by thedetermination unit 6 of the focustarget determination unit 4 . At this time, thedistance measuring unit 7 specifies a subject to focus on based on the two-dimensional coordinates input from the determiningunit 6 .

例如，测距部7构成为通过向被摄物体射出红外线等来对到被摄物体的距离进行测距，从而取得焦点信息。在该情况下，测距部7根据从For example, thedistance measuring unit 7 is configured to measure the distance to the subject by emitting infrared rays or the like toward the subject, thereby obtaining focus information. In this case, the rangingsection 7 according to the

〔〔输入部〕〕[[Input Section]]

输入部11使用快门释放按钮等来构成。当输入部11检测到用户的命令的输入时，根据所输入的命令，对摄像装置1的各部进行命令通知。例如，当输入部11检测到用户的自动聚焦的命令的输入时，向摄像部2通知自动聚焦的命令。并且，例如，当输入部11检测到用户的摄像命令的输入时，向摄像部2通知摄像的命令。Theinput unit 11 is configured using a shutter release button or the like. When theinput unit 11 detects the input of a command from the user, it notifies each unit of theimaging device 1 of the command based on the input command. For example, when theinput unit 11 detects the user's input of an auto-focus command, it notifies theimaging unit 2 of the auto-focus command. Furthermore, for example, when theinput unit 11 detects the user's input of an imaging command, it notifies theimaging unit 2 of the imaging command.

在输入部11使用快门释放按钮来构成的情况下，自动聚焦命令通过例如将该快门释放按钮半按下，由输入部11来进行检测。并且，摄像命令通过例如将该快门释放按钮完全按下，由输入部11来进行检测。When theinput unit 11 is configured using a shutter release button, the autofocus command is detected by theinput unit 11 by pressing the shutter release button halfway, for example. Then, an imaging command is detected by theinput unit 11 by, for example, fully pressing the shutter release button.

〔动作例〕〔Example of action〕

图5是表示摄像装置1的动作例的流程图。以下，使用图5对摄像装置1的动作例进行说明。FIG. 5 is a flowchart showing an example of the operation of theimaging device 1 . Hereinafter, an operation example of theimaging device 1 will be described using FIG. 5 .

当摄像装置1接通电源时，摄像部2把聚焦镜头控制成成为基于规定的焦点信息的状态。When theimaging device 1 is powered on, theimaging unit 2 controls the focus lens to be in a state based on predetermined focus information.

当输入部11检测到快门释放按钮被用户半按下时(S01)，把输入了自动聚焦命令的情况通知给摄像部2。When theinput unit 11 detects that the shutter release button has been pressed halfway by the user ( S01 ), it notifies theimaging unit 2 that an autofocus command has been input.

摄像部2判断是否从测距部7输入了焦点信息。在未输入焦点信息的情况下(S02-否)，摄像部2在聚焦镜头被控制成基于规定的焦点信息的状态下进行摄像(S03)。摄像部2把所摄像的图像数据输出给对焦对象决定部4。Theimaging unit 2 determines whether or not focus information has been input from thedistance measuring unit 7 . When the focus information is not input (S02-No), theimaging unit 2 performs imaging while the focus lens is controlled based on predetermined focus information (S03). Theimaging unit 2 outputs the captured image data to the focustarget determination unit 4 .

对焦对象决定部4根据所输入的图像数据，决定进行对焦的被摄物体(S04)。图6是表示对焦对象决定部4的动作例的流程图。使用图6对S04的处理进行说明。The focustarget determination unit 4 determines a subject to focus on based on the input image data ( S04 ). FIG. 6 is a flowchart showing an example of the operation of the focustarget determination unit 4 . The processing of S04 will be described using FIG. 6 .

首先，脸检测部5从所输入的图像中检测人物的脸(S10)。脸检测部5把所检测出的脸的图像信息输出给决定部6。First, theface detection unit 5 detects a person's face from an input image (S10). Theface detection unit 5 outputs the image information of the detected face to thedetermination unit 6 .

决定部6调查由脸检测部5所检测出的脸的数目。当脸的数目小于1的情况下(S11：<1)，即为0的情况下，决定部6把画面中央的二维坐标输出给测距部7。在脸的数目大于1的情况下(S11：>1)，即在检测出多个脸的情况下，决定部6执行第一处理，从所检测出的多个脸中选择一个脸(S13)。Thedetermination unit 6 checks the number of faces detected by theface detection unit 5 . When the number of faces is less than 1 (S11: <1), ie, 0, thedetermination unit 6 outputs the two-dimensional coordinates of the center of the screen to thedistance measuring unit 7 . When the number of faces is greater than 1 (S11:>1), that is, when a plurality of faces are detected, thedecision section 6 executes a first process to select one face from the detected faces (S13) .

在所检测出的脸是一个的情况下(S11：＝1)，或者在S13的处理后，决定部6根据一个脸来决定进行对焦的被摄物体(S14)。这里所说的一个脸在所检测出的脸的数目是一个的情况下是该唯一被检测出的脸，在S13处理后的情况下是被选择的脸。并且，决定部6取得所决定的被摄物体的二维坐标，把该二维坐标输出给测距部7(S15)。When there is one detected face (S11:=1), or after the processing of S13, thedetermination unit 6 determines the subject to focus on based on one face (S14). The one face mentioned here is the only detected face when the number of detected faces is one, and is the selected face after the processing in S13. Then, the determiningunit 6 acquires the determined two-dimensional coordinates of the subject, and outputs the two-dimensional coordinates to the distance measuring unit 7 ( S15 ).

回到使用图5的说明。在S04的处理后，测距部7根据从对焦对象决定部4的决定部6中输出的二维坐标，取得针对所决定的被摄物体的焦点信息(S05)。测距部7把所取得的焦点信息输出给摄像部2。在该处理后，再次执行S02及之后的处理。Returning to the description using FIG. 5 . After the process of S04 , thedistance measuring unit 7 acquires focus information on the determined subject based on the two-dimensional coordinates output from thedetermination unit 6 of the focus target determination unit 4 ( S05 ). Thedistance measuring unit 7 outputs the acquired focus information to theimaging unit 2 . After this processing, the processing in and after S02 is executed again.

在S02中，在从测距部7到摄像部2有焦点信息的输入的情况下(S02-是)，摄像部2根据所输入的焦点信息来控制聚焦镜头(S06)。即，摄像部2控制聚焦镜头，以便对由对焦对象决定部4所决定的被摄物体对焦。In S02, when there is focus information input from thedistance measuring unit 7 to the imaging unit 2 (S02-Yes), theimaging unit 2 controls the focus lens based on the input focus information (S06). That is, theimaging unit 2 controls the focus lens so as to focus on the subject determined by the focustarget determination unit 4 .

在聚焦镜头的控制后，摄像部2进行摄像(S07)。通过该摄像，摄像对由对焦对象决定部4所决定的被摄物体进行了对焦的图像。并且，图像存储部3存储由摄像部2所摄像的图像的数据(S08)。After controlling the focus lens, theimaging unit 2 performs imaging (S07). Through this imaging, an image focused on the subject determined by the in-focustarget determination unit 4 is captured. Then, theimage storage unit 3 stores data of the image captured by the imaging unit 2 (S08).

〔作用/效果〕〔Effect〕

图7是表示由摄像装置1所摄像的图像的示例的图。在摄像装置1中，检测出作为主要被摄物体的脸(由粗线条矩形围起来的脸)，而不是作为背景而被摄像的人物的脸，以对该主要被摄物体对焦的方式来实施摄像。FIG. 7 is a diagram showing an example of an image captured by theimaging device 1 . In theimaging device 1, the face (the face surrounded by a thick rectangle) is detected as the main subject, not the face of the person photographed as the background, and the main subject is captured by focusing on it. .

在摄像装置1中，对焦对象决定部4的脸检测部5检测人物的脸作为成为对焦对象的被摄物体。此时，脸检测部5根据关注区域内的第一特征区域和第二特征区域的浓淡差进行脸检测。在获取某区域内的浓淡值时，即使在没有对该区域内所摄像的被摄物体对焦的情况下，也能获得接近于进行了对焦的情况下的值。因此，在摄像装置1中，即使在摄像了焦点没有对准人物的脸的图像的情况下，也能检测该脸。因此，在摄像装置1中，在摄像检测人物的脸时所使用的图像时，没有必要进行对被摄物体的焦点控制。因此，在该摄像处理时，在摄像装置1中，没有必要针对被控制为规定焦点的摄像部2的聚焦镜头重新进行焦点控制，可快速地执行摄像。In theimaging device 1 , theface detection unit 5 of the focustarget determination unit 4 detects the face of a person as a subject to be focused. At this time, theface detection unit 5 performs face detection based on the shade difference between the first characteristic region and the second characteristic region in the region of interest. When acquiring the shading value in a certain area, even when the subject captured in the area is not in focus, a value close to that in the case of focusing can be obtained. Therefore, in theimaging device 1 , even when an image out of focus on a person's face is captured, the face can be detected. Therefore, in theimaging device 1 , it is not necessary to perform focus control on a subject when imaging an image used for detecting a person's face. Therefore, in this imaging process, in theimaging device 1 , it is not necessary to re-execute focus control on the focus lens of theimaging unit 2 controlled to a predetermined focus, and imaging can be performed quickly.

并且，对焦对象决定部4仅针对一个图像，即被控制为规定焦点而摄像的图像进行脸检测。因此，在摄像装置1中，没有必要对多个图像进行脸检测，可快速地检测脸。Furthermore, the focustarget determination unit 4 performs face detection on only one image, that is, an image controlled to be captured at a predetermined focus. Therefore, in theimaging device 1, it is not necessary to perform face detection on a plurality of images, and a face can be detected quickly.

并且，在摄像装置1中，在检测出多个脸的情况下，对焦对象决定部4选择任意一个脸，根据该脸决定成为对焦对象的被摄物体。因此，根据摄像装置1，用户没有必要特意从被收纳在帧内的多个脸中选择一个脸。Furthermore, in theimaging device 1 , when a plurality of faces are detected, the focustarget determination unit 4 selects any one face, and determines a subject to be focused based on the face. Therefore, according to theimaging device 1 , the user does not need to deliberately select one face from the plurality of faces stored in the frame.

〔变形例〕〔Modification〕

对焦对象决定部4可以独立构成为摄像装置1的外部装置，而不包括在摄像装置1内。并且，这样构成为外部装置的对焦对象决定部4可以构成为在自身的装置内配备具有与测距部7相同功能的未作图示的测距部。The focustarget determination unit 4 may be independently configured as an external device of theimaging device 1 instead of being included in theimaging device 1 . Furthermore, the focusingtarget determining unit 4 configured as an external device in this way may be configured to include a distance measuring unit (not shown) having the same function as thedistance measuring unit 7 in its own device.

另外，摄像装置1通过使用测距部7对到被摄物体的距离进行测距来取得焦点信息，然而也可以构成为使用测距以外的方法来取得焦点信息。In addition, theimaging device 1 obtains focus information by measuring the distance to the subject using thedistance measuring unit 7 , but may be configured to obtain focus information using a method other than distance measurement.

另外，摄像装置1也可以构成为根据所检测出的脸决定在实施摄像时成为进行曝光控制的对象的被摄物体，由此来进行曝光控制。在该情况下，决定部6决定成为进行曝光控制的对象的被摄物体。并且，在该情况下，摄像装置1还具有测光部，该测光部根据由决定部6所决定的被摄物体取得曝光控制信息。测光部通过把由决定部6所决定的被摄物体作为曝光区域进行测光来算出曝光量(曝光控制信息)。然后，测光部把所算出的曝光控制信息传递到摄像部2。然后，摄像部2通过调节光圈等来实施基于该曝光控制信息的曝光控制，进行摄像。In addition, theimaging device 1 may be configured to perform exposure control by determining a subject to be subjected to exposure control when imaging is performed based on a detected face. In this case, thedetermination unit 6 determines a subject to be subjected to exposure control. In addition, in this case, theimaging device 1 further includes a photometry unit that acquires exposure control information based on the subject determined by thedetermination unit 6 . The photometry unit calculates the exposure amount (exposure control information) by performing photometry with the subject determined by thedetermination unit 6 as an exposure area. Then, the photometry unit transmits the calculated exposure control information to theimaging unit 2 . Then, theimaging unit 2 performs exposure control based on the exposure control information by adjusting the aperture or the like, and performs imaging.

另外，脸检测部5也可以构成为把具有肤色的被摄物体检测为脸。通过这样构成，尽管不能准确地检测人物的脸，然而可削减检测处理所需要的时间，也能削减脸检测部5的安装所需要的时间和成本。In addition, theface detection unit 5 may be configured to detect a subject having a skin color as a face. With this configuration, although the face of a person cannot be accurately detected, the time required for the detection process can be reduced, and the time and cost required for mounting theface detection unit 5 can also be reduced.

另外，摄像部2也可以作为手动聚焦装置来实现，而不是自动聚焦功能。在该情况下，可以构成为例如当由用户操作了聚焦镜头，成为适合于从测距部7输入的焦点信息的状态时，把已进行了对焦的情况报知给用户。In addition, theimaging unit 2 may be implemented as a manual focus device instead of an auto focus function. In this case, for example, when the focus lens is operated by the user to be in a state suitable for the focus information input from thedistance measuring unit 7 , it may be configured to notify the user that the focus has been achieved.

另外，摄像装置1也可以构成为摄像到胶卷上的银铅照相机，而不是使用具有CCD的摄像部2来最终实施摄像。在该情况下，摄像装置1具有具备CCD的摄像部2以及作为银铅照相机的摄像部。并且，在该情况下，图像存储部3对摄像装置1而言并不是必要的。摄像部2为了把预备摄像的图像输入到对焦对象决定部4中而实施摄像。并且，在作为银铅照相机的摄像部中实施测距部7的基于焦点信息的聚焦镜头的调节。然而，摄像部2和作为银铅照相机的摄像部可以构成为共用例如聚焦镜头等其结构的一部分。In addition, theimaging device 1 may be configured as a silver-lead camera that captures images on film, instead of using theimaging unit 2 having a CCD to finally perform imaging. In this case, theimaging device 1 has animaging unit 2 including a CCD and an imaging unit that is a silver-lead camera. Also, in this case, theimage storage unit 3 is not necessary for theimaging device 1 . Theimaging unit 2 performs imaging in order to input the preliminarily captured image to the focustarget determination unit 4 . In addition, the adjustment of the focus lens based on the focus information of thedistance measuring unit 7 is carried out in the imaging unit which is a silver-lead camera. However, theimaging unit 2 and the imaging unit that is a silver-lead camera may be configured to share a part of their configuration, for example, a focusing lens.

[第二～第六实施方式的概要][Summary of Second to Sixth Embodiments]

以下，使用图对具有对焦对象决定部16(16a、16b、16c、16d、16e)的摄像装置12(12a、12b、12c、12d、12e)进行说明。在该说明中，人物图像是至少包含人物的脸的一部分或全部的图像的图像。因此，人物图像可以包含人物整体的图像，也可以包含仅人物的脸或仅上半身的图像。并且，可以包含针对多个人物的图像。而且，背景可以包含人物以外的风景(背景：也包含作为被摄物体而被关注的物体)和图案等任何图形。Hereinafter, the imaging device 12 (12a, 12b, 12c, 12d, 12e) provided with the in-focus target determination part 16 (16a, 16b, 16c, 16d, 16e) is demonstrated using a figure. In this description, a person image is an image including at least part or all of a person's face. Therefore, the image of a person may include an image of the entire person, or an image of only the face or upper body of the person. Also, images for multiple persons may be included. Furthermore, the background may include any figures such as scenery (background: also includes objects of interest as subjects) and patterns other than people.

另外，针对对焦对象决定部16和摄像装置12的以下说明仅是例示，其结构不限于以下说明。In addition, the following description about the focustarget determination part 16 and the imaging device 12 is an example only, and the structure is not limited to the following description.

[第二实施方式][Second Embodiment]

〔系统结构〕〔system structure〕

首先，对具有对焦对象决定部16a的摄像装置12a进行说明。摄像装置12a在硬件上具有通过总线连接的CPU(中央运算处理装置)、主存储装置(RAM)、辅助存储装置以及用于作为数字相机和数字摄像机进行动作的各装置(摄像镜头、机械机构、CCD、操作部、马达等)等。辅助存储装置使用非易失性存储装置来构成。这里所说的非易失性存储装置是指所谓的ROM、FRAM、硬盘等。First, theimaging device 12a having the focustarget determination unit 16a will be described. Theimaging device 12a has a CPU (central processing unit) connected via a bus, a main storage device (RAM), an auxiliary storage device, and various devices for operating as a digital camera and a digital video camera (imaging lens, mechanical mechanism, CCD, operating unit, motor, etc.) etc. The auxiliary storage device is configured using a nonvolatile storage device. The nonvolatile storage device referred to here refers to so-called ROM, FRAM, hard disk, and the like.

图8是表示摄像装置12a的功能块的图。摄像装置12a通过把存储在辅助存储装置内的各种程序(OS、应用程序等)加载到主存储装置内并使用CPU来执行，从而作为包括输入部13、摄像部14、图像存储部15、对焦对象决定部16a以及测距部21等的装置来发挥功能。对焦对象决定部16a是通过使用CPU来执行对焦对象决定程序而实现的。另外，对焦对象决定部16a也可以作为专用硬件来实现。FIG. 8 is a diagram showing functional blocks of theimaging device 12a. Theimaging device 12a loads various programs (OS, application programs, etc.) stored in the auxiliary storage device into the main storage device and executes them using the CPU, thereby operating as aninput unit 13, animaging unit 14, animage storage unit 15, Devices such as the focustarget determination unit 16a and thedistance measuring unit 21 function. The focustarget determination unit 16a is realized by executing a focus target determination program using a CPU. In addition, the in-focustarget determination unit 16a may be realized as dedicated hardware.

以下，使用图8对摄像装置12a内所包含的各功能部进行说明。Hereinafter, each functional unit included in theimaging device 12a will be described using FIG. 8 .

〈输入部〉〈Input part〉

输入部13使用快门释放按钮等来构成。当输入部13检测到用户的命令的输入时，根据所输入的命令，对摄像装置12a的各部进行命令通知。例如，当输入部13检测到用户的自动聚焦命令的输入时，向摄像部14通知自动聚焦命令。并且，例如，当输入部13检测到用户的摄像命令的输入时，向摄像部14通知摄像命令。Theinput unit 13 is configured using a shutter release button or the like. When theinput unit 13 detects the input of a command from the user, it notifies each unit of theimaging device 12 a of the command based on the input command. For example, when theinput unit 13 detects an input of an auto focus command by the user, it notifies theimaging unit 14 of the auto focus command. Furthermore, for example, when theinput unit 13 detects the user's input of an imaging command, it notifies theimaging unit 14 of the imaging command.

在输入部13使用快门释放按钮来构成的情况下，自动聚焦命令是通过例如将该快门释放按钮半按下，由输入部13来进行检测。并且，摄像命令是通过例如将该快门释放按钮完全按下，由输入部13来进行检测。When theinput unit 13 is configured using a shutter release button, the autofocus command is detected by theinput unit 13 by pressing the shutter release button halfway, for example. In addition, the imaging command is detected by theinput unit 13 by, for example, fully pressing the shutter release button.

〈摄像部〉〈camera department〉

摄像部14使用摄像镜头、机械机构、CCD、马达等，构成为具有自动聚焦功能的装置。摄像镜头包括例如用于实现变焦功能的变焦镜头、用于对任意被摄物体进行对焦的聚焦镜头等。机械机构包括机械快门、光圈、滤光器等。马达包括变焦镜头马达、聚焦马达、快门马达等。Theimaging unit 14 is configured as a device having an autofocus function using an imaging lens, a mechanism, a CCD, a motor, and the like. The imaging lens includes, for example, a zoom lens for realizing a zoom function, a focus lens for focusing on an arbitrary subject, and the like. Mechanical mechanisms include mechanical shutters, apertures, filters, etc. The motor includes a zoom lens motor, a focus motor, a shutter motor, and the like.

摄像部14具有未作图示的存储部，把规定的焦点信息存储在该存储部内。规定的焦点信息是预先决定的一个焦点信息。在摄像装置12a的电源接通的情况下，如果从测距部21没有焦点信息的输入，则摄像部14的聚焦镜头被控制成成为基于该规定的焦点信息的状态。另一方面，如果从测距部21有焦点信息的输入，则摄像部14的聚焦镜头被控制成成为基于所输入的焦点信息的状态。Theimaging unit 14 has a storage unit (not shown), and stores predetermined focus information in the storage unit. The prescribed focus information is a piece of focus information determined in advance. When theimaging device 12a is powered on, if there is no focus information input from thedistance measuring unit 21, the focus lens of theimaging unit 14 is controlled to be in a state based on the predetermined focus information. On the other hand, when focus information is input from thedistance measuring unit 21 , the focus lens of theimaging unit 14 is controlled to be in a state based on the input focus information.

摄像部14通过使用CCD把经由包含有像上述那样被控制的聚焦镜头的摄像镜头而成像的被摄物体的图像转换成电信号，来进行摄像。Theimaging unit 14 performs imaging by using a CCD to convert an image of a subject formed through an imaging lens including a focus lens controlled as described above into an electrical signal.

摄像部14可以构成为每当用户通过输入部13进行了输入时，就摄像1张或多张静态图像。另外，摄像部14也可以构成为根据通过输入部13的用户输入，摄像由多个帧构成的动态图像(时序图像)。另外，摄像部14在构成为摄像动态图像的情况下，可以构成为从多个帧中取得一个帧的图像，把该帧的图像作为处理对象的图像传递给对焦对象决定部16a。Theimaging unit 14 may be configured to capture one or more still images every time the user inputs through theinput unit 13 . In addition, theimaging unit 14 may be configured to capture a moving image (time-series image) composed of a plurality of frames in response to user input via theinput unit 13 . In addition, when theimaging unit 14 is configured to capture a moving image, it may be configured to obtain an image of one frame out of a plurality of frames, and pass the image of the frame to the focustarget determination unit 16a as an image to be processed.

〈图像存储部〉<Image storage unit>

图像存储部15使用所谓的RAM等可读写记录介质等来构成。图像存储部15也可以使用相对于摄像装置12a可拆装的记录介质来构成。图像存储部15存储由摄像部14所摄像的图像的数据。Theimage storage unit 15 is configured using a readable and writable recording medium such as a so-called RAM. Theimage storage unit 15 may also be configured using a recording medium that is detachable from theimaging device 12a. Theimage storage unit 15 stores data of images captured by theimaging unit 14 .

〈对焦对象决定部〉<Focus Target Determination Unit>

对焦对象决定部16a从由摄像部14所摄像的被摄物体中决定进行对焦的被摄物体，即成为对焦对象的脸。此时，对焦对象决定部16a使用由摄像部14为了对焦用而摄像的图像来进行处理。该为了对焦用而摄像的图像由摄像部14输入给对焦对象决定部16a。The focustarget determination unit 16 a determines a subject to be focused on, that is, a face to be a focus target, from among subjects captured by theimaging unit 14 . At this time, the in-focustarget determination unit 16 a performs processing using an image captured by theimaging unit 14 for focusing. The image captured for focusing is input from theimaging unit 14 to the focustarget determination unit 16 a.

对焦对象决定部16a包括脸检测部17a和决定部18a。以下，对构成对焦对象决定部16a的各功能部进行说明。The focustarget determination unit 16a includes aface detection unit 17a and adetermination unit 18a. Hereinafter, each functional unit constituting the in-focustarget determination unit 16a will be described.

〈脸检测部〉〈Face detection part〉

脸检测部17a从对焦用图像中获得包含人物的脸的区域(以下称为“脸区域”)的坐标。根据该脸矩形的坐标，确定对焦用图像中的人的脸区域的位置和大小。Theface detection unit 17 a obtains coordinates of a region including a person's face (hereinafter referred to as a "face region") from the focusing image. Based on the coordinates of the face rectangle, the position and size of the face area of the person in the focusing image are determined.

脸区域的检测可以通过应用现有的任何技术来实现。例如，可以通过使用与脸整体的轮廓对应的基准模板的模板匹配来获得脸矩形的坐标。另外，也可以通过基于脸的构成要素(眼、鼻、耳等)的模板匹配来获得脸矩形的坐标。另外，也可以通过色度键处理来检测头发的顶点，根据该顶点获得脸矩形的坐标。另外，也可以根据肤色区域的检测获得脸矩形的坐标。Detection of face regions can be achieved by applying any existing techniques. For example, the coordinates of the face rectangle can be obtained by template matching using a reference template corresponding to the outline of the entire face. Alternatively, the coordinates of the face rectangle may be obtained by template matching based on face components (eyes, nose, ears, etc.). Alternatively, the vertex of the hair may be detected by chroma key processing, and the coordinates of the face rectangle may be obtained from the vertex. In addition, the coordinates of the face rectangle can also be obtained according to the detection of the skin color area.

〈决定部〉〈decision department〉

决定部18a从由脸检测部17a所检测出的脸中决定成为对焦对象的脸。决定部18a把与所决定的脸有关的脸矩形的坐标(例如表示脸矩形的中央的坐标)传递给测距部21。决定部18a包括中心决定部19a和脸决定部20a。以下，对构成决定部18a的各功能部进行说明。Thedetermination unit 18a determines a face to be focused on from the faces detected by theface detection unit 17a. The determiningunit 18 a transmits the coordinates of the face rectangle related to the determined face (for example, coordinates indicating the center of the face rectangle) to thedistance measuring unit 21 . Thedetermination unit 18a includes a center determination unit 19a and a face determination unit 20a. Hereinafter, each functional unit constituting thedetermination unit 18a will be described.

〈中心决定部〉〈Central Decision Department〉

中心决定部19a在由脸检测部17a所检测出的脸是多个的情况下，求出所检测出的多个脸的位置的中心。这里，假定脸矩形的中央点(以下称为“脸点”)表示脸的位置。在所检测出的脸是二个的情况下，中心决定部19a求出二个脸点的中点。并且，在所检测出的脸是三个以上的情况下，中心决定部19a使用以下任一种方法求出中心。The center determination unit 19 a obtains the centers of the positions of the plurality of detected faces when there are a plurality of faces detected by theface detection unit 17 a. Here, it is assumed that the central point of the face rectangle (hereinafter referred to as "face point") indicates the position of the face. When there are two detected faces, the center determination unit 19a finds the midpoint of the two face points. In addition, when there are three or more detected faces, the center determination unit 19a obtains the center using any of the following methods.

〈〈外接多边形法〉〉<<Circumscribed Polygon Method>>

在外接多边形法中，中心决定部19a求出与多个脸点外接的多边形的中心作为多个脸的位置的中心。所用的多边形的顶点数无论是多少个均可以，也可以预先决定。这里，对顶点数是4点的情况进行说明。In the circumscribing polygon method, the center determination unit 19 a obtains the centers of polygons circumscribing the plurality of face points as the centers of the positions of the plurality of faces. The number of vertices of the polygon to be used may be any number, and may be determined in advance. Here, a case where the number of vertices is four will be described.

图9是用于对中心决定部19a根据外接多边形法取得中心的处理进行说明的图。中心决定部19a分别针对多个脸点的x坐标和y坐标，判断最大值和最小值。中心决定部19a生成由与y轴平行的直线和与x轴平行的直线所生成的矩形(在正交坐标系中，由与x轴和y轴平行且分别通过脸点坐标的各坐标成分的最大值和最小值的各条线的交点构成顶点的矩形)，该与y轴平行的直线包含取x坐标的最大值和最小值的脸点，该与x轴平行的直线包含取y坐标的最大值和最小值的脸点。并且，中心决定部19a取得所生成的矩形的对角线的交点的坐标作为中心坐标。FIG. 9 is a diagram for explaining a process of obtaining a center by the center determination unit 19 a by the circumscribed polygon method. The center determining unit 19a determines the maximum value and the minimum value for each of the x-coordinates and y-coordinates of a plurality of face points. The center determination unit 19a generates a rectangle formed by a straight line parallel to the y-axis and a straight line parallel to the x-axis (in the orthogonal coordinate system, each coordinate component parallel to the x-axis and y-axis and passing through the face point coordinates respectively The intersection of each line of maximum value and minimum value constitutes the rectangle of the apex), the straight line parallel to the y axis contains the face point with the maximum value and minimum value of the x coordinate, the straight line parallel to the x axis contains the point of the y coordinate Face points for maximum and minimum values. Then, the center determination unit 19a acquires the coordinates of the intersection of the diagonals of the generated rectangle as the center coordinates.

〈〈重心法〉〉<<Center of gravity method>>

图10是用于对中心决定部19a根据重心法取得中心的处理进行说明的图。在重心法中，中心决定部19a取得多个脸点的重心。具体地说，中心决定部19a通过将各脸点的位置矢量的和除以脸点的数目来取得重心的位置矢量，根据该位置矢量取得重心坐标。然后，中心决定部19a把所取得的重心坐标取作中心坐标。图10示出了针对6个脸点，分别求出距某基准点(例如利用各脸点的坐标分量各自的最小值来表示的点)的位置矢量，求出重心坐标(Xg、Yg)的状况。FIG. 10 is a diagram for explaining the process of obtaining the center by the center determination unit 19 a by the center of gravity method. In the center of gravity method, the center determination unit 19a acquires the centers of gravity of a plurality of face points. Specifically, the center determination unit 19a obtains the position vector of the center of gravity by dividing the sum of the position vectors of the face points by the number of face points, and obtains the coordinates of the center of gravity from the position vector. Then, the center determination unit 19a takes the obtained center-of-gravity coordinates as the center coordinates. Fig. 10 shows, for 6 face points, respectively obtain the position vector from a certain reference point (for example, the point represented by the respective minimum values of the coordinate components of each face point), and obtain the barycentric coordinates (Xg, Yg) situation.

〈脸决定部〉〈Face decision department〉

脸决定部20a根据由中心决定部19a所取得的中心坐标，决定可以根据哪个脸进行对焦。图11是用于对脸决定部20a的处理进行说明的图。脸决定部20a根据规定基准从位于距中心规定距离内的脸中选择脸。此时，脸决定部20a根据脸的大小、脸的位置、脸的朝向等中的哪一个基准来决定脸均可。例如，脸决定部20a在位于距由中心决定部19a所取得的中心规定距离内的脸中选择距图像(帧)的中央点(图11中由符号“×”图示)最近的脸(由粗线条圆圈围起来的脸点)。The face determining unit 20a determines which face can be focused based on the center coordinates acquired by the center determining unit 19a. FIG. 11 is a diagram for explaining the processing of the face determination unit 20a. The face determination unit 20a selects a face from faces located within a predetermined distance from the center based on a predetermined criterion. At this time, the face determination unit 20a may determine the face based on any reference among the size of the face, the position of the face, and the orientation of the face. For example, the face determination unit 20a selects the face (indicated by the symbol “×” in FIG. 11 ) closest to the center point of the image (frame) among the faces located within a predetermined distance from the center acquired by the center determination unit 19a. face points surrounded by thick line circles).

在图11所示的例中，针对在以距脸点的中心F0规定距离R为半径的圆C1内存在的二个脸点F1和F2，测定与图像(帧)的中心点○的距离L1和L2。然后，由于L1<L2，因而把脸点F1决定为对焦对象的脸点(即对焦对象的脸)。In the example shown in FIG. 11, for two face points F1 and F2 existing in a circle C1 having a predetermined distance R from the center F0 of the face point as a radius, the distance L1 from the center point ○ of the image (frame) is measured. and L2. Then, since L1<L2, the face point F1 is determined as the face point of the focus object (ie, the face of the focus object).

脸决定部20a把所选择的脸点的坐标传递给测距部21。The face determining unit 20 a transmits the coordinates of the selected face point to thedistance measuring unit 21 .

〈测距部〉〈Distance measuring department〉

测距部21取得用于对由对焦对象决定部16a的脸决定部20a所决定的脸对焦的焦点信息。此时，测距部21根据从脸决定部20a输入的二维坐标(脸点坐标)，确定用于进行对焦的被摄物体。由测距部21所实施的测距可以应用例如通过向脸的实物射出红外线来实施的方式(有源方式)，也可以应用这种有源方式以外的方式(例如无源方式)。Thedistance measuring unit 21 acquires focus information for focusing on the face determined by the face determination unit 20a of the focustarget determination unit 16a. At this time, thedistance measuring unit 21 specifies a subject to focus on based on the two-dimensional coordinates (face point coordinates) input from the face determining unit 20 a. For the distance measurement by thedistance measuring unit 21 , for example, a method (active method) by emitting infrared rays toward the actual face may be applied, or a method other than this active method (for example, a passive method) may be applied.

例如，测距部21构成为通过向被摄物体射出红外线等对到被摄物体的距离进行测距，从而取得焦点信息。在该情况下，测距部21根据从脸决定部20a输入的二维坐标，决定射出红外线等的方向。For example, thedistance measuring unit 21 is configured to measure the distance to the subject by emitting infrared rays or the like toward the subject, thereby obtaining focus information. In this case, thedistance measuring unit 21 determines the direction in which infrared rays and the like are emitted based on the two-dimensional coordinates input from the face determination unit 20a.

当测距部21取得焦点信息时，把所取得的焦点信息传递给摄像部14。When thedistance measuring unit 21 acquires focus information, it passes the acquired focus information to theimaging unit 14 .

〔动作例〕〔Example of action〕

图12是表示摄像装置12a的动作例的流程图。以下，使用图12对摄像装置12a的动作例进行说明。FIG. 12 is a flowchart showing an example of the operation of theimaging device 12a. Hereinafter, an operation example of theimaging device 12a will be described using FIG. 12 .

当摄像装置12a接通电源时，摄像部14把聚焦镜头控制成成为基于规定的焦点信息的状态。When theimaging device 12a is powered on, theimaging unit 14 controls the focus lens to be in a state based on predetermined focus information.

当输入部13检测到快门释放按钮被用户半按下时(S16)，把输入了自动聚焦命令的情况通知给摄像部14。When theinput unit 13 detects that the shutter release button has been pressed halfway by the user ( S16 ), it notifies theimaging unit 14 that an autofocus command has been input.

摄像部14判断从测距部21是否输入了焦点信息。在未输入焦点信息的情况下(S17-否)，摄像部14摄像对焦用图像(S18)。摄像部14把所摄像的对焦用图像传递给对焦对象决定部16a。Theimaging unit 14 determines whether or not focus information has been input from thedistance measuring unit 21 . When the focus information is not input (S17-No), theimaging unit 14 captures an image for focusing (S18). Theimaging unit 14 transmits the captured focusing image to the focusingtarget determination unit 16a.

对焦对象决定部16a根据所输入的对焦用图像，决定成为进行对焦的对象的脸(S19)。图13是表示对焦对象决定部16a的动作例(进行对焦的脸的决定处理(S19)的例子)的流程图。使用图13对S19的处理进行说明。The focustarget determination unit 16 a determines a face to be focused on based on the input focus image ( S19 ). FIG. 13 is a flowchart showing an example of the operation of the in-focustarget determination unit 16a (an example of the process of determining the face to be in focus (S19)). The processing of S19 will be described using FIG. 13 .

首先，脸检测部17a从所输入的图像中检测人物的脸(S24)。脸检测部17a把所检测出的脸的图像信息传递给决定部18a的中心决定部19a。First, theface detection unit 17a detects a person's face from an input image (S24). Theface detection unit 17a transmits the image information of the detected face to the center determination unit 19a of thedetermination unit 18a.

中心决定部19a调查由脸检测部17a所检测出的脸的数目。当脸的数目小于1的情况下(S25：<1)，即为0的情况下，中心决定部19a把没有脸的情况通知给脸决定部20。并且，脸决定部20a把图像中央的二维坐标传递给测距部21(S26)。The center determination unit 19a checks the number of faces detected by theface detection unit 17a. When the number of faces is less than 1 (S25: <1), ie, 0, the center determination unit 19a notifies theface determination unit 20 that there are no faces. Then, the face determining unit 20a transmits the two-dimensional coordinates of the center of the image to the distance measuring unit 21 (S26).

在脸的数目多于1的情况下(S25：>1)，即在检测出多个脸的情况下，决定部18a执行从多个脸中选择一个脸的处理(S27)。图14是示出从多个脸中选择一个脸的处理(S27)的例子的流程图。使用图14对S27的处理进行说明。When the number of faces is more than 1 (S25:>1), that is, when a plurality of faces are detected, thedetermination unit 18a executes a process of selecting one of the plurality of faces (S27). FIG. 14 is a flowchart showing an example of a process of selecting one face from a plurality of faces (S27). The processing of S27 will be described using FIG. 14 .

中心决定部19a取得所检测出的脸的脸点的中心坐标(S29)。中心决定部19a把所取得的中心坐标传递给脸决定部20a。然后，脸决定部20a根据所输入的中心坐标，采用例如使用图11所说明的方法选择一个脸(S30)。The center determination unit 19a acquires the center coordinates of the face points of the detected face (S29). The center determination unit 19a transmits the acquired center coordinates to the face determination unit 20a. Then, the face determination unit 20a selects one face by, for example, the method described using FIG. 11 based on the input center coordinates (S30).

回到使用图13的说明。在从图像中所检测出的脸是一个的情况下(S25：＝1)，或者在S27的处理后，脸决定部20a把一个脸的脸点的二维坐标传递给测距部21(S28)。这里所说的一个脸，在所检测出的脸的数目是一个的情况下是该唯一被检测出的脸，在S27的处理后的情况下是被选择的脸。Returning to the description using FIG. 13 . In the case of one face detected from the image (S25:=1), or after the processing of S27, the face determination part 20a transmits the two-dimensional coordinates of the face point of one face to the distance measuring part 21 (S28 ). The one face mentioned here is the only detected face when the number of detected faces is one, and is the selected face after the process of S27.

回到使用图12的说明。在S19的处理后，测距部21根据从对焦对象决定部16a的脸决定部20a所传递的二维坐标取得焦点信息(S20)。测距部21把所取得的焦点信息传递给摄像部14。在该处理后，再次执行S17及之后的处理。Returning to the description using FIG. 12 . After the process of S19, thedistance measuring unit 21 acquires focus information based on the two-dimensional coordinates passed from the face determination unit 20a of the focustarget determination unit 16a (S20). Thedistance measuring unit 21 transmits the acquired focus information to theimaging unit 14 . After this processing, the processing in and after S17 is executed again.

在S17中，在从测距部21到摄像部14有焦点信息的输入的情况下(S17-是)，摄像部14根据所输入的焦点信息控制聚焦镜头(S21)。即，摄像部14控制聚焦镜头，以便对由对焦对象决定部16a所决定的脸对焦。In S17, when there is input of focus information from thedistance measuring unit 21 to the imaging unit 14 (S17-Yes), theimaging unit 14 controls the focus lens based on the input focus information (S21). That is, theimaging unit 14 controls the focus lens so as to focus on the face determined by the focustarget determination unit 16a.

在聚焦镜头的控制后，摄像部14进行摄像(S22)。通过该摄像，摄像了对由对焦对象决定部16a所决定的脸进行了对焦的图像。然后，图像存储部15存储由摄像部14所摄像的图像的数据(S23)。After the control of the focus lens, theimaging unit 14 performs imaging (S22). Through this imaging, an image in which the face determined by the in-focustarget determination unit 16 a is in focus is captured. Then, theimage storage unit 15 stores data of the image captured by the imaging unit 14 (S23).

〔作用/效果〕〔Effect〕

根据作为本发明的第二实施方式的摄像装置12a，检测图像中存在的人脸，自动把所检测出的脸中的任意一个决定为对焦对象。因此，用户没有必要使要成为对焦对象的脸位于图像(帧)的中心。换句话说，用户没有必要把要对焦的脸放置在帧的中心。According to theimaging device 12a which is the second embodiment of the present invention, human faces existing in an image are detected, and any one of the detected faces is automatically determined as a focus target. Therefore, it is not necessary for the user to position the face to be the focus target at the center of the image (frame). In other words, it is not necessary for the user to place the face to be in focus in the center of the frame.

并且，根据作为本发明的第二实施方式的摄像装置12a，在图像中存在多个脸的情况下，根据图像中存在的多个脸的位置的中心，决定成为对焦对象的脸。因此，即使在图像中存在多个脸的情况下，也能决定应成为对焦对象的一个脸。并且，可以对位于集合中的多个脸的中央附近的脸进行对焦。Furthermore, according to theimaging device 12a according to the second embodiment of the present invention, when there are multiple faces in the image, the face to be focused on is determined based on the centers of the positions of the multiple faces in the image. Therefore, even when there are multiple faces in the image, one face to be focused on can be determined. Also, it is possible to focus on a face located near the center of a plurality of faces in the set.

〔变形例〕〔Modification〕

脸决定部20a可以构成为根据规定基准，从位于距中心规定距离内的且朝向正面的脸中选择脸。此时，脸决定部20a使用例如以下公知文献所描述的技术来判断脸是否朝向正面。The face determination unit 20a may be configured to select a face from faces facing the front within a predetermined distance from the center based on a predetermined criterion. At this time, the face determination unit 20a determines whether the face is facing the front using, for example, the technique described in the following known documents.

H.Schneiderman，T.Kanade.“A Statistical Method for 3DObject Detection Applied to Faces and Cars.”IEEE Conference onComputer Vision and Pattern Recognition，2000.H. Schneiderman, T. Kanade. "A Statistical Method for 3DObject Detection Applied to Faces and Cars." IEEE Conference on Computer Vision and Pattern Recognition, 2000.

并且，在上述描述中对中心决定部19a把脸的位置作为点来获取的Computer Vision and Pattern Recognition，2000.Also, in the above description, the center determination unit 19a acquires the position of the face as a point in Computer Vision and Pattern Recognition, 2000.

并且，在上述描述中对中心决定部19a把脸的位置作为点来获取的结构进行了说明，然而中心决定部19a也可以构成为把脸的位置作为区域来获取。Furthermore, in the above description, the structure in which the center determination unit 19a acquires the position of the face as a point has been described, but the center determination unit 19a may be configured to acquire the position of the face as a region.

并且，也可以构成为由脸决定部20a所决定的脸通过未作图示的显示部，在该脸的周围显示框。通过这样构成，用户可知道哪个脸被决定为对焦对象。并且，可以设计成在用户不满意对这样被显示的脸进行对焦的情况下，通过使用输入部13输入命令，可以通过手动操作以便对其它脸对焦。In addition, a frame may be displayed around the face determined by the face determination unit 20 a through a display unit (not shown). With this configuration, the user can know which face is determined as the focus target. In addition, when the user is dissatisfied with focusing on the face displayed in this way, by inputting a command using theinput unit 13, it is possible to manually operate to focus on another face.

并且，脸决定部20a可以构成为根据规定基准，从位于距中心规定距离内、且其中的规定大小以上的脸中选择脸。In addition, the face determination unit 20a may be configured to select a face from among faces located within a predetermined distance from the center and having a predetermined size or larger, based on a predetermined criterion.

并且，摄像装置12a也可以构成为根据由决定部18a所决定的脸，执行曝光控制。这点在后述的第二至第六实施方式中也相同。Furthermore, theimaging device 12a may be configured to perform exposure control based on the face determined by thedetermination unit 18a. This point is also the same in the second to sixth embodiments described later.

并且，决定部18a也可以将由脸检测部17a所检测出的各脸根据其x坐标或y坐标进行排序，把与所检测出的脸的数目的中间值的顺序相当的脸决定为成为对象的脸。In addition, thedetermination unit 18a may sort the faces detected by theface detection unit 17a according to their x-coordinates or y-coordinates, and determine a face corresponding to the order of the median value of the number of detected faces as the target face. Face.

并且，决定部18a也可以构成为根据规定基准从由脸检测部17a所检测出的脸中决定脸。规定基准例如可以是基于脸的大小的基准(例如决定最大的脸的基准，决定中间大小的脸的基准，决定最小的脸的基准)，可以是基于图像中的脸的位置的基准(例如距图像中央最近的脸的基准)，可以是基于脸的朝向的基准(例如朝向正面的脸的基准)，可以是基于类似于脸的基准(例如决定表示类似于脸的程度的量为最大的脸的基准)，可以是基于根据脸而推测的性别的基准(例如决定被推测为男性的脸的基准，决定被推测为女性的脸的基准)，可以是基于根据脸而推测的年代的基准(决定被推测为最年轻的年代的脸的基准，决定被推测为中间年代的脸的基准)，是其它哪种基准均可。并且，规定基准也可以是将上述基准适当地进行多个组合的基准。Furthermore, thedetermination unit 18a may be configured to determine a face from the faces detected by theface detection unit 17a based on a predetermined criterion. The predetermined criterion may be, for example, a criterion based on the size of the face (for example, a criterion for determining the largest face, a criterion for determining a middle-sized face, and a criterion for determining the smallest face), or a criterion based on the position of the face in the image (for example, a distance from The reference of the nearest face in the center of the image), may be a reference based on the orientation of the face (for example, a reference for a face facing the front), may be a reference based on a similar face (for example, the face whose degree of similarity to a face is determined to be the largest ), which may be based on the gender estimated from the face (for example, the basis for determining a face that is estimated to be male, the basis for determining a face that is estimated to be female), or the basis based on the age estimated from the face ( The reference for determining the face estimated to be the youngest age, the reference for determining the face estimated to be the middle age), or any other reference may be used. Furthermore, the predetermined criterion may be a combination of a plurality of the above-mentioned criteria as appropriate.

并且，在上述说明中，前提是摄像部14构成为对一个被摄物体进行对焦。因此，对焦对象决定部16a把一个脸的坐标传递给测距部21。然为这种摄像部14的例子，有特开平11-295826所描述的装置。在该情况下，决定部5a的脸决定部20a根据规定基准对脸赋予优先顺序，根据该优先顺序选择规定数目的脸。例如，脸决定部20a可以构成为按照距由中心决定部19a所决定的中心从近到远的顺序选择规定数目的脸。在所选择的脸的数目未达到规定数的情况下，使用其它哪种方法来取得传递给测距部21的坐标均可。In addition, in the above description, it is assumed that theimaging unit 14 is configured to focus on one subject. Therefore, the focustarget determining unit 16 a transmits the coordinates of one face to thedistance measuring unit 21 . However, as an example of such animaging unit 14, there is a device described in JP-A-11-295826. In this case, the face determination unit 20 a of the determination unit 5 a prioritizes the faces based on a predetermined criterion, and selects a predetermined number of faces based on the priority. For example, the face determination unit 20a may be configured to select a predetermined number of faces in order from the closest to the center determined by the center determination unit 19a. When the number of selected faces does not reach the predetermined number, any other method may be used to acquire the coordinates to be passed to thedistance measuring unit 21 .

[第三实施方式][Third Embodiment]

下面，对具有对焦对象决定部16b的摄像装置12b进行说明。以下，针对摄像装置12b和对焦对象决定部16b，说明与第二实施方式中的摄像装置12a和对焦对象决定部16a的不同点。Next, the imaging device 12b having the in-focus target determination unit 16b will be described. Hereinafter, differences from theimaging device 12a and the focusingtarget determining unit 16a in the second embodiment will be described with respect to the imaging device 12b and the focusing target determining unit 16b.

〔系统结构〕〔system structure〕

图15是示出摄像装置12b的功能块的图。取代对焦对象决定部16a，摄像装置12b具有对焦对象决定部16b，这点与第二实施方式中的摄像装置12a不同。因此，以下，省略对摄像装置12b的输入部13、摄像部14、图像存储部15以及测距部21的说明。FIG. 15 is a diagram showing functional blocks of the imaging device 12b. The imaging device 12b is different from theimaging device 12a in the second embodiment in that it includes a focusing target determining unit 16b instead of the focusingtarget determining unit 16a. Therefore, descriptions of theinput unit 13 , theimaging unit 14 , theimage storage unit 15 , and thedistance measuring unit 21 of the imaging device 12 b will be omitted below.

〈对焦对象决定部〉<Focus Target Determination Unit>

对焦对象决定部16b包括脸检测部17a、分类部22、集合决定部23以及决定部18b。以下，对构成对焦对象决定部16b的各功能部进行说明。然而，由于对焦对象决定部16b所包括的脸检测部17a具有与第二实施方式中的对焦对象决定部16a所包括的脸检测部17a相同的结构，因而省略其说明。The focus target determination unit 16b includes aface detection unit 17a, aclassification unit 22, anaggregation determination unit 23, and a determination unit 18b. Hereinafter, each functional unit constituting the in-focus target determination unit 16b will be described. However, since theface detection unit 17a included in the focus target determination unit 16b has the same configuration as theface detection unit 17a included in the focustarget determination unit 16a in the second embodiment, description thereof will be omitted.

〈分类部〉〈Classification Department〉

分类部22把在脸检测部17a中所检测出的脸的脸点分类为多个集合。该分类可以使用最邻近法等任何方法。分类部22根据各脸点的位置执行分类。分类部22的分类结果是将各脸点分类成使得图像中的位置相近的脸点被包含在同一集合内。Theclassification unit 22 classifies the face points of the faces detected by theface detection unit 17a into a plurality of sets. This classification can use any method such as the nearest neighbor method. Theclassification unit 22 performs classification based on the position of each face point. As a result of the classification by theclassification unit 22 , the face points are classified so that face points that are close in position in the image are included in the same set.

图16是示出分类部22的处理例的图。如图16所示，在存在多个脸点的集体的情况下，把每一个集体分类为一个集合。FIG. 16 is a diagram illustrating an example of processing by theclassification unit 22 . As shown in FIG. 16, in the case where there are a plurality of collections of face points, each collection is classified into a set.

〈集合决定部〉〈Assembly Decision Department〉

集合决定部23决定在分类部22中分类的多个集合中的用于决定成为对焦对象的脸的集合(即包含成为对焦对象的脸的集合：以下称为“选择集合”)。即，传递给测距部21的脸点坐标是由集合决定部23所决定的选择集合内所包含的脸的脸点坐标。Theset determination unit 23 determines a set for determining a face to be in focus (that is, a set including the face to be in focus: hereinafter referred to as a “selection set”) among the sets classified by theclassification unit 22 . That is, the face point coordinates passed to thedistance measuring unit 21 are the face point coordinates of the faces included in the selection set determined by theset determination unit 23 .

集合决定部23根据各集合内所包含的脸的数目、与各集合内所包含的脸有关的特征量等哪种基准来决定选择集合均可。例如，集合决定部23把各集合中包含最多的脸的集合决定为选择集合。并且，例如，集合决定部23把与各集合内所包含的脸有关的特征量的和最大的集合决定为选择集合。与脸有关的特征量可以根据例如脸的大小、脸的朝向、脸的数目等来取得。并且，与脸有关的特征量可以是表示人脸程度的值。人脸程度例如可以根据距用于划分某图像是否是人脸的识别边界线的距离来给出。人脸程度也可以使用在脸检测部17a的脸检测处理时所取得的值来表示。Theset determination unit 23 may determine the selection set based on any criteria such as the number of faces included in each set, feature amounts related to faces included in each set, or the like. For example, theset determination unit 23 determines a set including the most faces among the sets as a selection set. Furthermore, for example, theset determination unit 23 determines the set with the largest sum of the feature values of the faces included in each set as the selected set. The feature amount related to a face can be obtained from, for example, the size of the face, the orientation of the face, the number of faces, and the like. Also, the feature amount related to a face may be a value indicating the degree of a human face. The degree of human face can be given, for example, according to the distance from the recognition boundary line used to classify whether an image is a human face or not. The degree of a human face can also be expressed using a value obtained during the face detection processing of theface detection unit 17a.

使用图16对集合决定部23的处理例进行说明。根据规定基准(这里是集合内所包含的脸点数多的基准)，通过集合决定部23从通过分类部22的处理所生成的多个集合(图16中为三个集合)中选择左上的集合作为选择集合。An example of processing by theset determination unit 23 will be described with reference to FIG. 16 . Based on a predetermined criterion (here, the criterion that the number of face points included in the set is large), theset determination unit 23 selects the upper left set from the plurality of sets (three sets in FIG. 16 ) generated by the processing of theclassification unit 22. as a selection set.

〈决定部〉〈decision department〉

决定部18b不是根据由摄像部14所摄像的图像整体内所包含的脸，而是根据由集合决定部23所决定的选择集合内所包含的脸进行处理(把选择集合作为脸的决定的母集合)，这点与决定部18a不同。决定部18b除了上述点以外，进行与决定部18a相同的处理。The determining unit 18b performs processing based not on the faces included in the entire image captured by theimaging unit 14, but on the basis of the faces included in the selected set determined by the set determining unit 23 (using the selected set as the mother of the determination of the face). set), which is different from thedetermination unit 18a. The decision unit 18b performs the same process as thedecision unit 18a except for the above points.

图17是示出决定部18b的处理例的图。通过集合决定部23决定的选择集合中，通过决定部18b的中心决定部19b取得脸点的中心。然后，根据所取得的中心，通过脸决定部20b，根据规定基准(这里是距图像(帧)的中央点近的基准)选择一个脸点(由粗线条圆圈围起来的脸点)。FIG. 17 is a diagram illustrating an example of processing by the determination unit 18b. In the selection set determined by theset determination unit 23, the center of the face point is acquired by the center determination unit 19b of the determination unit 18b. Then, based on the acquired center, a face point (a face point surrounded by a thick line circle) is selected by the face determination unit 20b based on a predetermined reference (here, a reference near the center point of the image (frame)).

〔动作例〕〔Example of action〕

下面，对第三实施方式中的摄像装置12b的动作例进行说明。摄像装置12b的动作除了S27的处理内容以外，与第二实施方式中的摄像装置12a的动作相同。因此，关于摄像装置12b的动作例，仅对S27的处理内容进行说明。Next, an operation example of the imaging device 12b in the third embodiment will be described. The operation of the imaging device 12b is the same as that of theimaging device 12a in the second embodiment except for the processing content of S27. Therefore, only the processing content of S27 will be described regarding the operation example of the imaging device 12b.

图18是示出第三实施方式中的对焦对象决定部16b的处理的一部分，即图13中的S27的处理内容的流程图。在第三实施方式中，在S27中，执行该图18所示的处理来取代在第二实施方式中所说明的图14的处理。FIG. 18 is a flowchart showing a part of the processing of the focus target determination unit 16 b in the third embodiment, that is, the processing content of S27 in FIG. 13 . In the third embodiment, in S27, the processing shown in FIG. 18 is executed instead of the processing in FIG. 14 described in the second embodiment.

当开始S27的处理时，分类部22把由脸检测部17a所检测出的多个脸点分类为多个集合(S31)。然后，集合决定部23从由分类部22所生成的多个集合中选择一个选择集合(S32)。When the process of S27 is started, theclassification unit 22 classifies the plurality of face points detected by theface detection unit 17a into a plurality of sets (S31). Then, theset determination unit 23 selects one selected set from the plurality of sets generated by the classification unit 22 ( S32 ).

中心决定部19b根据在S32中所选择的选择集合内所包含的脸，取得该选择集合内所包含的多个脸点的中心(S33)。然后，脸决定部20b根据在S33中所取得的中心，选择一个脸作为成为对焦对象的脸(S34)。The center determination unit 19b obtains the centers of the plurality of face points included in the selection set selected in S32 based on the faces included in the selection set (S33). Then, the face determination unit 20b selects one face as a face to be focused on based on the center acquired in S33 (S34).

〔作用/效果〕〔Effect〕

根据第三实施方式中的摄像装置12b，在图像中存在多个集体的情况下，可以仅根据其中一个集体内所包含的人，决定成为对焦对象的脸。在用户摄像某集体的情况下，多数情况是该集体的人集中在一个场所。此时，即使不属于该集体的人(单独人、单独人的零散分布、其它人的集体)存在于帧内，也期望从属于用户想要摄像的集体的人中选择成为对焦对象的被摄物体。根据第三实施方式中的摄像装置12b，可以实现用户的这种期望。即，在第三实施方式中，汇集在帧内的数量最多的人的集体(集合)被选择为摄像对象的集体(选择集合)，实现以属于该集体的任何人为基准的对焦。另外，如后面的变形例栏所述，可以构成为在由集合决定部23所选择的选择集合不是用户期望的集合的情况下，可由用户手动选择任意集合作为选择集合。According to the imaging device 12b in the third embodiment, when there are a plurality of groups in an image, it is possible to determine a face to be focused on based only on persons included in one of the groups. When a user takes pictures of a certain group, the people of the group gather in one place in many cases. At this time, even if people who do not belong to the group (a single person, a scattered distribution of single people, or a group of other people) exist in the frame, it is desirable to select a subject to be focused from among people belonging to the group that the user wants to capture. object. According to the imaging device 12b in the third embodiment, such a desire of the user can be fulfilled. That is, in the third embodiment, the group (collection) of the largest number of people gathered in the frame is selected as the group of imaging targets (selection group), and focusing based on any person belonging to the group is realized. In addition, as described in the section of Modifications below, when the selection set selected by theset determination unit 23 is not the set desired by the user, an arbitrary set can be manually selected by the user as the selection set.

图19是用于表示由摄像装置12b所取得的效果的图。如图19所示，即使在存在多个集体的情况下，也仅根据例如位于图像左下方的集体所包含的脸的位置，选择距该集体的中心近的脸(由粗线条矩形围起来的脸)。FIG. 19 is a diagram illustrating effects obtained by the imaging device 12b. As shown in FIG. 19, even when there are a plurality of groups, only the faces included in the group located at the lower left of the image are selected based on the positions of the faces near the center of the group (faces surrounded by thick-lined rectangles, for example). ).

〔变形例〕〔Modification〕

集合决定部23也可以构成为根据各集合内所包含的脸的脸点的中心来决定选择集合。在该情况下，中心决定部19b针对由分类部22所生成的各集合内所包含的脸点取得中心。集合决定部23根据由中心决定部19b所取得的各集合的中心，决定选择集合。例如，集合决定部23把集合内所包含的脸点的中心坐标距图像的中央点最近的集合决定为选择集合。并且，脸决定部20b根据选择集合的中心，决定成为对焦对象的脸。Theset determination unit 23 may be configured to determine the selection set based on the center of the face points of the faces included in each set. In this case, the center determination unit 19 b acquires the centers of the face points included in each set generated by theclassification unit 22 . Theset determination unit 23 determines a selection set based on the centers of the sets acquired by the center determination unit 19b. For example, theset determination unit 23 determines the set in which the center coordinates of the face points included in the set is closest to the center point of the image as the selected set. Then, the face determination unit 20b determines a face to be focused on based on the center of the selection set.

并且，集合决定部23也可以构成为根据在各集合中被暂定决定为成为对焦对象的脸的脸的脸点，决定选择集合。在该情况下，决定部18b针对由分类部22所生成的各集合，仅根据该集合内所包含的脸，暂定决定成为对焦对象的脸。并且，集合决定部23根据由决定部18b所取得的各集合中的脸点，决定选择集合。例如，集合决定部23把集合内所包含的被暂定决定的脸点的坐标距图像的中央点最近的集合决定为选择集合。即，实质上集合决定部23最终决定成为对焦对象的脸。在该情况下，集合决定部23也可以构成为把所决定的脸的脸点坐标传递给测距部21。In addition, theset determination unit 23 may be configured to determine the selection set based on the face points of the face tentatively determined as the face to be focused in each set. In this case, the determination unit 18 b tentatively determines, for each set generated by theclassification unit 22 , the face to be focused on based only on the faces included in the set. Then, theset determination unit 23 determines a selection set based on the face points in each set acquired by the determination unit 18b. For example, theset determination unit 23 determines a set in which the coordinates of tentatively determined face points included in the set are closest to the center point of the image as the selected set. That is, in essence, thegroup determination unit 23 finally determines the face to be focused on. In this case, theset determination unit 23 may be configured to pass the determined face point coordinates of the face to thedistance measuring unit 21 .

并且，在这样构成的情况下，也可以构成为在由集合决定部23所选择的集合不是用户期望的集合的情况下(集合的选择结果通过未作图示的显示部被提示给用户)，由用户使用输入部13选择其它集合作为选择集合。在该情况下，可以构成为每当输入部13被操作时，由用户选择的集合就根据优先顺序而移位。该优先顺序是基于由集合决定部23所判断的规定基准的顺序。In addition, in the case of such a configuration, when the group selected by thegroup determination unit 23 is not the group desired by the user (the selection result of the group is presented to the user by the display unit not shown), Other sets are selected by the user using theinput unit 13 as the selected set. In this case, the set selected by the user may be shifted according to the priority order every time theinput unit 13 is operated. This priority order is an order based on a predetermined criterion determined by theset determination unit 23 .

如上所述，针对每个集合暂定决定对焦对象的脸(有时也称为“对象脸”)，从暂定的对象脸中决定最终对象脸，把该对象脸提示给用户，在与用户的意图不一致的情况下，可以把对象脸变更为被暂定决定的其它对象脸中的任意一个。As described above, for each set, tentatively determine the face of the focus target (sometimes referred to as "target face"), determine the final target face from the tentative target faces, and present the target face to the user. If the intention is inconsistent, the target face can be changed to any one of other tentatively determined target faces.

取代上述构成，可以构成为在决定部18b从暂定对象脸中决定最终对象脸之前，把暂定对象脸提示给用户，由用户从其中选择最终对象脸。暂定对象脸通过未作图示的显示部，在该脸的周围显示框来提示给用户。Instead of the above-mentioned configuration, before the determination unit 18b determines the final target face from among the tentative target faces, it may be configured to present tentative target faces to the user and allow the user to select the final target face among them. The provisional target face is presented to the user by displaying a frame around the face on a display unit (not shown).

或者，可以在决定对象脸之前，把分类部22的集合分类结果通过未作图示的显示部提示给用户，由用户决定选择集合，决定部18b从由用户所决定的选择集合中选择对象脸。在该情况下，集合决定部23不进行选择集合的决定处理，而仅进行把用户的选择集合的决定结果传递给决定部18b的动作。这样，可以通过选项操作来打开/关闭集合决定部23的决定动作。Alternatively, before determining the target face, the set classification result of theclassification unit 22 may be presented to the user through a display unit not shown in the figure, and the user determines the selection set, and the determination unit 18b selects the target face from the selection set determined by the user. . In this case, theset determining unit 23 does not perform the process of determining the selected set, and only performs an operation of notifying the determination result of the user's selected set to the determining unit 18b. In this way, the determination operation of theaggregate determination unit 23 can be turned on/off by an option operation.

[第四实施方式][Fourth embodiment]

下面，对具有对焦对象决定部16c的摄像装置12c进行说明。以下，针对摄像装置12c和对焦对象决定部16c，说明与第三实施方式中的摄像装置12b和对焦对象决定部16b的不同点。Next, theimaging device 12c including the focustarget determination unit 16c will be described. Hereinafter, differences from the imaging device 12b and the focusing target determining unit 16b in the third embodiment will be described with respect to theimaging device 12c and the focusingtarget determining unit 16c.

〔系统结构〕〔system structure〕

图20是表示摄像装置12c的功能块的图。摄像装置12c具有对焦对象决定部16c来取代对焦对象决定部16b，这点与第三实施方式中的摄像装置12b不同。因此，以下，省略对摄像装置12c的输入部13、摄像部14、图像存储部15以及测距部21的说明。FIG. 20 is a diagram showing functional blocks of theimaging device 12c. Theimaging device 12c is different from the imaging device 12b in the third embodiment in that it has a focustarget determination unit 16c instead of the focus target determination unit 16b. Therefore, descriptions of theinput unit 13 , theimaging unit 14 , theimage storage unit 15 , and thedistance measuring unit 21 of theimaging device 12 c will be omitted below.

〈对焦对象决定部〉<Focus Target Determination Unit>

对焦对象决定部16c包括决定部18c来取代决定部18b，这点与第三实施方式中的对焦对象决定部16b不同。因此，省略对脸检测部17a、分类部22以及集合决定部23的说明。The focustarget determination unit 16c is different from the focus target determination unit 16b in the third embodiment in that it includes a determination unit 18c instead of the determination unit 18b. Therefore, descriptions of theface detection unit 17a, theclassification unit 22, and thegroup determination unit 23 are omitted.

〈决定部〉〈decision department〉

决定部18c根据由集合决定部23所决定的集合(选择集合)内所包含的脸进行处理。决定部18c根据该脸的大小，决定成为对焦对象的脸。决定部18c把例如选择集合内所包含的脸中的最大的脸决定为成为对焦对象的脸。或者，决定部18c也可以把例如由集合决定部23所决定的选择集合内所包含的脸中的具有中间大小的脸决定为成为对焦对象的脸。在以下说明中，设决定部18c把最大的脸决定为成为对焦对象的脸来进行说明。The decision unit 18c performs processing based on the faces included in the set (selection set) determined by theset decision unit 23 . The determination unit 18c determines a face to be focused on based on the size of the face. The determination unit 18 c determines, for example, the largest face among the faces included in the selection set as the face to be focused on. Alternatively, the determination unit 18 c may determine, for example, a face with an intermediate size among the faces included in the selection set determined by theset determination unit 23 as the face to be focused on. In the following description, it will be described assuming that the determination unit 18c determines the largest face as the face to be focused on.

〔动作例〕〔Example of action〕

下面，对第四实施方式中的摄像装置12c的动作例进行说明。摄像装置12c的动作除了S27的处理内容以外，与第三实施方式中的摄像装置12b的动作相同。因此，关于摄像装置12c的动作例，仅对S27的处理内容进行说明。Next, an operation example of theimaging device 12c in the fourth embodiment will be described. The operation of theimaging device 12c is the same as that of the imaging device 12b in the third embodiment except for the processing content of S27. Therefore, as an example of the operation of theimaging device 12c, only the processing content of S27 will be described.

图21是示出第四实施方式中的对焦对象决定部16c的处理的一部分，即图13中的S27的处理内容的流程图。FIG. 21 is a flowchart showing a part of the processing of the focustarget determination unit 16 c in the fourth embodiment, that is, the processing content of S27 in FIG. 13 .

当开始S27的处理时，分类部22把由脸检测部17a所检测出的多个脸点分类为多个集合(S31)。然后，集合决定部23从由分类部22所生成的多个集合中选择一个选择集合(S32)。并且，决定部18c根据各脸的大小，从在S32的处理中所选择的选择集合内所包含的脸中选择一个脸作为成为对焦对象的脸(S35)。When the process of S27 is started, theclassification unit 22 classifies the plurality of face points detected by theface detection unit 17a into a plurality of sets (S31). Then, theset determination unit 23 selects one selected set from the plurality of sets generated by the classification unit 22 ( S32 ). Then, the determination unit 18 c selects one face as the face to be focused on from the faces included in the selection set selected in the process of S32 according to the size of each face ( S35 ).

〔作用/效果〕〔Effect〕

根据第四实施方式中的摄像装置12c，与第三实施方式中的摄像装置12b一样，在图像中存在多个集体的情况下，可以仅根据其中一个集合内所包含的人，决定成为对焦对象的脸。According to theimaging device 12c in the fourth embodiment, similar to the imaging device 12b in the third embodiment, when there are multiple groups in the image, it is possible to determine the focus target based on only the people included in one of the groups. s face.

并且，根据第四实施方式中的摄像装置12c，从所选择的一个集合内所包含的脸中，根据该脸的大小，选择成为对焦对象的脸。有时，对于用户而言最想关注的脸是距摄像装置最近的脸。并且，距摄像装置最近的脸在摄像时是最大的脸的可能性高。因此，根据摄像装置12c，从所选择的一个集合内所包含的脸中，根据脸的大小选择被设想为对用户而言最想关注的脸。因此，在该情况下，用户没有必要手动地对想要关注的脸对焦。至少，没有必要像以往技术那样，把最想关注的脸放置在帧的中心或其附近。Furthermore, according to theimaging device 12c in the fourth embodiment, a face to be focused on is selected from among the faces included in one selected set according to the size of the face. Sometimes, the face that the user most wants to pay attention to is the face closest to the camera. In addition, there is a high possibility that the face closest to the imaging device will be the largest face at the time of imaging. Therefore, thecamera 12c selects, from the faces included in one of the selected sets, the face that is supposed to be the most interesting for the user according to the size of the face. Therefore, in this case, it is not necessary for the user to manually focus on a face to be focused on. At the very least, there is no need to place the face of most interest at or near the center of the frame, as in previous techniques.

〔变形例〕〔Modification〕

决定部18c也可以构成为根据与由集合决定部23所决定的集合内所包含的各脸有关的特征量来进行处理。决定部18c取得与各脸有关的特征量。决定部18c构成为例如根据脸的大小、脸的朝向等，取得与各脸有关的特征量。另外，决定部18c构成为例如针对各脸，取得表示类似于脸的程度的量作为特征量。“表示类似于脸的程度的量”使用例如在脸检测部17a的脸检测处理时所取得的值来表示。具体地说，当脸检测部17a构成为在判断某区域内是否包含有脸时算出与该区域内的图像的类似于脸有关的值的情况下，把与该类似于脸有关的值用作特征量。The determination unit 18 c may be configured to perform processing based on the feature amount of each face included in the set determined by theset determination unit 23 . The determination unit 18c acquires feature amounts related to each face. The determination unit 18c is configured to acquire feature amounts related to each face based on, for example, the size of the face, the orientation of the face, and the like. In addition, the determination unit 18 c is configured to acquire, for each face, a quantity indicating a degree of similarity to the face as a feature quantity, for example. "Amount indicating the degree of resemblance to a face" is represented by, for example, a value acquired during the face detection processing of theface detection unit 17a. Specifically, when theface detection unit 17a is configured to calculate a value related to a similar face of an image in a certain region when judging whether a face is included in the region, the value related to the similar face is used as Feature amount.

[第五实施方式][Fifth Embodiment]

下面，对具有对焦对象决定部16d的摄像装置12d进行说明。以下，针对摄像装置12d和对焦对象决定部16d，说明与第二实施方式中的摄像装置12a和对焦对象决定部16a的不同点。Next, theimaging device 12d having the in-focustarget determination unit 16d will be described. Hereinafter, differences from theimaging device 12a and the focusingtarget determining unit 16a in the second embodiment will be described with respect to theimaging device 12d and the focusingtarget determining unit 16d.

〔系统结构〕〔system structure〕

图22是表示摄像装置12d的功能块的图。摄像装置12d具有对焦对象决定部16d来取代对焦对象决定部16a，这点与第二实施方式中的摄像装置12a不同。因此，以下，省略对摄像装置12d的输入部13、摄像部14、图像存储部15以及测距部21的说明。FIG. 22 is a diagram showing functional blocks of theimaging device 12d. Theimaging device 12d is different from theimaging device 12a in the second embodiment in that it has a focustarget determination unit 16d instead of the focustarget determination unit 16a. Therefore, descriptions of theinput unit 13 , theimaging unit 14 , theimage storage unit 15 , and thedistance measuring unit 21 of theimaging device 12 d will be omitted below.

〈对焦对象决定部〉<Focus Target Determination Unit>

对焦对象决定部16d包括脸检测部17d、区决定部24以及决定部18d。以下，对构成对焦对象决定部16d的各功能部进行说明。The focustarget determination unit 16d includes aface detection unit 17d, anarea determination unit 24, and adetermination unit 18d. Hereinafter, each functional unit constituting the in-focustarget determination unit 16d will be described.

〈脸检测部〉〈Face detection part〉

脸检测部17d在从对焦用图像中检测脸区域时，针对预先决定的多个区的每一个检测脸区域，这点与第二实施方式中的脸检测部17a不同。图23是表示被分为多个区的对焦用图像的例子的图。如图23所示，对焦用图像被分为多个区(在图23中，矩形图像(帧)被与其纵或横轴平行且通过图像的中央点的二条线段分为4个区，然而分成多少个区均可)。例如由脸检测部17d存储将对焦用图像分为什么样的区。Theface detection unit 17d is different from theface detection unit 17a in the second embodiment in that it detects a face area for each of a plurality of predetermined areas when detecting a face area from the focusing image. FIG. 23 is a diagram showing an example of a focusing image divided into a plurality of areas. As shown in Figure 23, the image for focusing is divided into a plurality of areas (in Figure 23, a rectangular image (frame) is divided into four areas by two line segments parallel to its vertical or horizontal axis and passing through the central point of the image, but divided into as many districts as possible). For example, the area into which the focusing image is divided is stored in theface detection unit 17d.

脸检测部17d针对多个区的每个区检测脸。然后，脸检测部17d针对每一区把检测结果传递给区决定部24。Theface detection unit 17d detects faces for each of the plurality of areas. Then, theface detection unit 17d transmits the detection result to thearea determination unit 24 for each area.

〈区决定部〉〈City decision department〉

区决定部24选择预先决定的多个区中的用于决定成为对焦对象的脸的区(即，包含成为对焦对象的脸的区：以下称为“选择区”)。即，被传递给测距部21的坐标是在由区决定部24所决定的选择区中检测出的脸的脸点坐标。Thearea determination unit 24 selects an area for determining a face to be in focus (that is, an area including a face to be in focus: hereinafter referred to as a “selection area”) among a plurality of predetermined areas. That is, the coordinates passed to thedistance measuring unit 21 are face point coordinates of faces detected in the selection area determined by thearea determination unit 24 .

区决定部24根据在各区中所检测出的脸的数目、与在区中所检测出的脸有关的特征量等哪个基准来决定选择区均可。例如，区决定部24把在各区中检测出最多的脸的区决定为选择区。在这样构成区决定部24的情况下，在图23所示的状况下，左上部的区被区决定部24决定为选择区。Thearea determination unit 24 may determine the selection area based on any reference, such as the number of faces detected in each area, feature amounts related to faces detected in the area, or the like. For example, thearea determination unit 24 determines an area in which the most faces are detected among each area as a selection area. When thearea determination unit 24 is configured in this way, in the situation shown in FIG. 23 , the upper left area is determined by thearea determination unit 24 as a selection area.

并且，例如，区决定部24也可以把与各区内所包含的脸有关的特征量的和为最大的区决定为选择区。作为与脸有关的特征量的计算方法，可以应用与在上述实施方式中所说明的方法相同的方法。Furthermore, for example, thearea determination unit 24 may determine, as the selected area, the area in which the sum of the feature values related to the faces included in each area is the largest. As a calculation method of the feature amount related to a face, the same method as that described in the above-mentioned embodiment can be applied.

〈决定部〉〈decision department〉

决定部18d不是根据由摄像部14所摄像的图像整体内所包含的脸，而是根据由区决定部24所决定的选择区内所包含的脸来进行处理，这点与决定部18a不同。决定部18d除了上述点以外，进行与决定部18a相同的处理。The determiningunit 18d differs from the determiningunit 18a in that it performs processing based not on the faces included in the entire image captured by theimaging unit 14 but on the faces included in the selection area determined by thearea determining unit 24 . Thedecision unit 18d performs the same process as thedecision unit 18a except for the above points.

图24是表示决定部18d的处理例的图。在由区决定部24所选择的左上部的区(选择区)中，取得脸点的中心。然后，根据所取得的中心，按照规定基准(这里是距图像(帧)的中央点近的基准)，选择一个脸点(由粗线条圆圈围起来的脸点)。FIG. 24 is a diagram showing an example of processing by thedetermination unit 18d. In the upper left area (selected area) selected by thearea determination unit 24, the center of the face point is acquired. Then, based on the acquired center, a face point (a face point surrounded by a thick line circle) is selected according to a predetermined reference (here, a reference near the center point of the image (frame)).

〔动作例〕〔Example of action〕

下面，对第五实施方式中的摄像装置12d的动作例进行说明。摄像装置12d的动作除了S19的处理内容以外，与第二实施方式中的摄像装置12a的动作相同。因此，关于摄像装置12d的动作例，仅对S19的处理内容进行说明。Next, an operation example of theimaging device 12d in the fifth embodiment will be described. The operation of theimaging device 12d is the same as that of theimaging device 12a in the second embodiment except for the processing content of S19. Therefore, only the processing content of S19 will be described regarding the operation example of theimaging device 12d.

图25是示出第五实施方式中的对焦对象决定部16d的处理的一部分，即图12中的S19的处理内容的流程图。这样，在第五实施方式中，在S19中，取代在第二实施方式中所应用的图13所示的处理，而执行图25所示的处理。FIG. 25 is a flowchart showing a part of the processing of the focustarget determination unit 16d in the fifth embodiment, that is, the processing content of S19 in FIG. 12 . Thus, in the fifth embodiment, in S19, the processing shown in FIG. 25 is executed instead of the processing shown in FIG. 13 applied in the second embodiment.

当开始S19的处理时，脸检测部17d针对每个区检测脸(S36)。然后，区决定部24调查检测出脸的区的数目。在检测出脸的区的数目少于1的情况下(S37：<1)，即是0的情况下，区决定部24把没有脸的情况通知给决定部18d。然后，决定部18d把图像(帧)中央的二维坐标传递给测距部21(S38)。When the process of S19 starts, theface detection unit 17d detects a face for each area (S36). Then, thearea determination unit 24 checks the number of areas in which faces are detected. If the number of areas in which faces are detected is less than 1 (S37: <1), that is, if it is 0, thearea determination unit 24 notifies thedetermination unit 18d that there is no face. Then, thedetermination unit 18d transmits the two-dimensional coordinates of the center of the image (frame) to the distance measuring unit 21 (S38).

在检测出脸的区的数目多于1的情况下(S37：>1)，区决定部24从检测出脸的多个区中选择一个选择区(S39)。在检测出脸的区是一个的情况下(S37：＝1)，或者在S39的处理后，决定部18d调查由区决定部24所选择的选择区内所包含的脸的数目。在脸的数目多于1的情况下(S40：>1)，即在该选择区内检测出多个脸的情况下，决定部18d从该多个脸中选择一个脸(S41)。S41的处理是与图14所示的流程相同的处理。然而，处理对象不是图像整体而是选择区，这点与第二实施方式不同。When the number of face-detected areas is more than 1 (S37:>1), thearea determination unit 24 selects one selection area from the plurality of face-detected areas (S39). When there is only one face detection area (S37:=1), or after the process of S39, thedetermination unit 18d checks the number of faces included in the selection area selected by thearea determination unit 24. When the number of faces is more than 1 (S40:>1), that is, when a plurality of faces are detected in the selection area, thedetermination unit 18d selects one face from the plurality of faces (S41). The processing of S41 is the same processing as the flow shown in FIG. 14 . However, it is different from the second embodiment in that the object of processing is not the entire image but a selected area.

回到使用图25的说明。在所检测出的脸是一个的情况下(S40：＝1)，或者在S41的处理后，决定部18d把一个脸的脸点的二维坐标传递给测距部21(S42)。这里所说的一个脸，在所检测出的脸的数目是一个的情况下是该唯一被检测出的脸，在S41的处理后的情况下是被选择的脸。在该处理后，实施图12中的S20及之后的处理。Returning to the description using FIG. 25 . When there is one detected face (S40:=1), or after the process of S41, thedetermination unit 18d transmits the two-dimensional coordinates of the face point of one face to the distance measuring unit 21 (S42). The one face mentioned here is the only detected face when the number of detected faces is one, and is the selected face after the process of S41. After this processing, the processing in and after S20 in FIG. 12 is carried out.

〔作用/效果〕〔Effect〕

根据第五实施方式中的摄像装置12d，把对焦用图像分割成多个区，从多个区中选择包含成为对焦对象的脸的一个选择区。并且，从该一个选择区中选择成为对焦对象的脸。因此，在第五实施方式中的摄像装置12d中，可类似地取得与第三实施方式相同的效果。具体地说，在图像中存在多个集体的情况下，可使用区粗略划分这些集体。并且，根据按照规定基准所选择的一个选择区内所包含的集体或集体的一部分的脸，选择成为对焦对象的脸。According to theimaging device 12d in the fifth embodiment, the focusing image is divided into a plurality of regions, and one selection region including the face to be focused is selected from the plurality of regions. Then, a face to be focused is selected from the one selection area. Therefore, in theimaging device 12d in the fifth embodiment, the same effects as those of the third embodiment can be similarly obtained. Specifically, in the case where there are multiple collectives in an image, regions can be used to roughly divide these collectives. Then, a face to be focused is selected based on faces of a group or a part of a group included in one selection area selected according to a predetermined criterion.

并且，在第五实施方式中的摄像装置12d中，与第三实施方式中的摄像装置12b不同，不执行分类处理。因此，在第五实施方式中的摄像装置12d中，与第三实施方式中的摄像装置12b相比可进行快速处理。Also, in theimaging device 12d in the fifth embodiment, unlike the imaging device 12b in the third embodiment, classification processing is not performed. Therefore, in theimaging device 12d in the fifth embodiment, faster processing can be performed compared with the imaging device 12b in the third embodiment.

〔变形例〕〔Modification〕

决定部18d也可以与第三实施方式中的决定部18b相同。在该情况下，决定部18d构成为不是根据由集合决定部23所决定的选择集合内所包含的脸，而是根据由区决定部24所决定的选择区内所包含的脸进行处理，这点与决定部18b不同。即，构成为进行不是以图像整体而是以选择区为对象的集合的分类和选择集合的决定等。在该情况下，决定部18d构成为除了上述点以外，进行与决定部18b相同的处理。Thedetermination unit 18d may be the same as the determination unit 18b in the third embodiment. In this case, thedetermination unit 18d is configured to perform processing not based on the faces included in the selection set determined by theset determination unit 23, but based on the faces included in the selection area determined by thearea determination unit 24. The point is different from the determination unit 18b. That is, it is configured to perform classification of sets targeted at selected regions instead of entire images, determination of selected sets, and the like. In this case, thedetermination unit 18d is configured to perform the same processing as that of the determination unit 18b except for the above points.

通过这样构成，没有必要进行以画面内所包含的所有脸为对象的分类。即，只要仅对所选择的区内所包含的脸执行分类即可。因此，在画面内包含有较多个集体的情况下，与第三实施方式中的摄像装置12b相比可进行快速处理。并且，与通常的第五实施方式中的摄像装置12d相比，可更准确地选择包含成为对焦对象的脸的集合和成为对焦对象的脸。With such a configuration, it is not necessary to classify all faces included in the screen. That is, it is only necessary to classify only the faces included in the selected area. Therefore, when a large number of groups are included in the screen, faster processing can be performed compared with the imaging device 12b in the third embodiment. Furthermore, compared with theusual imaging device 12d of the fifth embodiment, it is possible to more accurately select a set including a face to be focused and a face to be focused.

[第六实施方式][Sixth Embodiment]

下面，对具有对焦对象决定部16e的摄像装置12e进行说明。以下，针对摄像装置12e和对焦对象决定部16e，说明与第二实施方式中的摄像装置12a和对焦对象决定部16a的不同点。Next, theimaging device 12e including the focus target determination unit 16e will be described. Hereinafter, differences from theimaging device 12a and the focusingtarget determining unit 16a in the second embodiment will be described with respect to theimaging device 12e and the focusing target determining unit 16e.

〔系统结构〕〔system structure〕

图26是表示摄像装置12e的功能区的图。摄像装置12e具有对焦对象决定部16e来取代对焦对象决定部16a，这点与第二实施方式中的摄像装置12a不同。因此，以下，省略对摄像装置12e的输入部13、摄像部14、图像存储部15以及测距部21的说明。FIG. 26 is a diagram showing functional areas of theimaging device 12e. Theimaging device 12e is different from theimaging device 12a in the second embodiment in that it has a focus target determination unit 16e instead of the focustarget determination unit 16a. Therefore, descriptions of theinput unit 13 , theimaging unit 14 , theimage storage unit 15 , and thedistance measuring unit 21 of theimaging device 12 e will be omitted below.

〈对焦对象决定部〉<Focus Target Determination Unit>

对焦对象决定部16e包括脸检测部17a、最大脸选择部25、候选选择部26以及决定部18e。以下，对构成对焦对象决定部16e的各功能部进行说明。其中，对焦对象决定部16e所包括的脸检测部17a与第二实施方式中的对焦对象决定部16a所包括的脸检测部17a结构相同，因而省略其说明。The focus target determination unit 16e includes aface detection unit 17a, a maximumface selection unit 25, acandidate selection unit 26, and adetermination unit 18e. Hereinafter, each functional unit constituting the in-focus target determination unit 16e will be described. Here, theface detection unit 17a included in the focus target determination unit 16e has the same configuration as theface detection unit 17a included in the focustarget determination unit 16a in the second embodiment, and thus its description is omitted.

〈最大脸选择部〉〈Maximum face selection part〉

最大脸选择部25从由脸检测部17a所检测出的多个脸中选择最大的脸(最大脸)。在脸检测部17a中使用脸矩形的情况下，最大脸选择部25根据该脸矩形的大小进行选择。并且，在脸检测部17a中执行图形匹配的情况下，最大脸选择部25根据该图形的大小进行选择。最大脸选择部25构成为使用其它哪种方法来选择最大脸均可。最大脸选择部25把关于所选择的最大脸的信息(例如脸点的坐标、脸的大小)传递给候选选择部26。The largestface selection unit 25 selects the largest face (largest face) from the plurality of faces detected by theface detection unit 17a. When a face rectangle is used in theface detection unit 17a, the maximumface selection unit 25 selects it according to the size of the face rectangle. Furthermore, when pattern matching is performed in theface detection unit 17a, the maximumface selection unit 25 selects according to the size of the pattern. The largestface selection unit 25 may be configured to select the largest face by any other method. The largestface selection section 25 passes information on the selected largest face (for example, the coordinates of face points, the size of the face) to thecandidate selection section 26 .

图27是示出最大脸选择部25的处理例的图。在图27中，各脸点的大小表示与各脸点对应的脸的大小。通过最大脸选择部25的处理，把由粗框围起来的脸点选择为最大脸。FIG. 27 is a diagram showing an example of processing by the largestface selection unit 25 . In FIG. 27 , the size of each face point indicates the size of the face corresponding to each face point. Through the processing of the largestface selection unit 25, the face point surrounded by a thick frame is selected as the largest face.

〈候选选择部〉〈Candidate selection department〉

候选选择部26根据由最大脸选择部25所选择的最大脸的大小，选择成为对焦对象的脸的候选。换句话说，候选选择部26决定成为决定部18e的处理对象的脸。Thecandidate selection unit 26 selects a candidate for a face to be focused on based on the size of the largest face selected by the largestface selection unit 25 . In other words, thecandidate selection unit 26 determines a face to be processed by thedetermination unit 18e.

候选选择部26把由最大脸选择部25所选择的最大脸、和具有基于该最大脸的大小的规定范围内的大小的其它脸选择为对象脸的候选。规定范围内的大小表示比最大脸小例如几个百分比～几十百分比的大小。另外，规定范围内的大小也表示例如为最大脸的一半以上的大小或2/3以上的大小。Thecandidate selection unit 26 selects the largest face selected by the largestface selection unit 25 and other faces having a size within a predetermined range based on the size of the largest face as candidates for the target face. The size within the predetermined range means a size smaller than the largest face, for example, several percent to several tens percent. In addition, the size within the predetermined range also means, for example, the size of half or more of the largest face, or the size of 2/3 or more.

图28是示出候选选择部26的处理例的图。在图28中，小的空心圆表示没有被候选选择部26选择的脸点。涂黑的圆表示由候选选择部26所选择的其它的脸的脸点。另外，由粗框包围的脸点表示由决定部18e所选择的最大脸的脸点。FIG. 28 is a diagram showing an example of processing by thecandidate selection unit 26 . In FIG. 28 , small hollow circles represent face points not selected by thecandidate selection unit 26 . Black circles indicate face points of other faces selected by thecandidate selection unit 26 . In addition, the face points surrounded by thick frames indicate the face points of the largest face selected by thedetermination unit 18e.

〈决定部〉〈decision department〉

决定部18e不是根据由摄像部14所摄像的图像整体内所包含的脸，而是根据由候选选择部26所选择的候选内所包含的脸进行处理，这点与决定部18a不同。决定部18e除了上述点以外，进行与决定部18a相同的处理。The determiningunit 18 e is different from the determiningunit 18 a in that it performs processing not based on the faces included in the entire image captured by theimaging unit 14 but based on the faces included in the candidates selected by thecandidate selecting unit 26 . Thedecision unit 18e performs the same process as thedecision unit 18a except for the above points.

使用图28对决定部18e的处理进行说明。决定部18e根据由候选选择部26所选择的三个脸点，选择成为对焦对象的脸。此时，决定部18e的中心决定部19e取得该三个脸点的中心(斜线的圆)。并且，决定部18e的脸决定部20e根据该中心，按照规定基准(这里是距图像(帧)的中心(中央点)最近的基准)选择一个脸点(这里是由粗框围起来的脸点(最大脸))。The processing of thedetermination unit 18e will be described using FIG. 28 . Thedetermination unit 18 e selects a face to be focused on based on the three face points selected by thecandidate selection unit 26 . At this time, the center determination unit 19e of thedetermination unit 18e obtains the centers (shaded circles) of the three face points. And, the face determination unit 20e of thedetermination unit 18e selects a face point (here, a face point surrounded by a thick frame) according to a predetermined reference (here, the reference closest to the center (central point) of the image (frame)) based on the center. (maximum face)).

〔动作例〕〔Example of action〕

下面，对第六实施方式中的摄像装置12e的动作例进行说明。摄像装置12e的动作除了S27的处理内容以外，与第二实施方式中的摄像装置12a的动作相同。因此，关于摄像装置12e的动作例，仅对S27的处理内容进行说明。Next, an example of the operation of theimaging device 12e in the sixth embodiment will be described. The operation of theimaging device 12e is the same as that of theimaging device 12a in the second embodiment except for the processing content of S27. Therefore, only the processing content of S27 is demonstrated about the operation example of theimaging device 12e.

图29是示出第六实施方式中的对焦对象决定部16e的处理的一部分，即图13中的S27的处理内容的流程图。这样，第六实施方式在S27中执行图29所示的处理来取代图14所示的处理，这点与第二实施方式不同。FIG. 29 is a flowchart showing a part of the processing of the focus target determination unit 16 e in the sixth embodiment, that is, the processing content of S27 in FIG. 13 . Thus, the sixth embodiment is different from the second embodiment in that the processing shown in FIG. 29 is executed in S27 instead of the processing shown in FIG. 14 .

当开始S27的处理时，最大脸选择部25选择由脸检测部17a所检测出的多个脸中的最大脸(S43)。然后，候选选择部26根据由最大脸选择部25所选择的脸(最大脸)的大小，选择成为候选的脸(S44)。When the process of S27 is started, the largestface selection unit 25 selects the largest face among the plurality of faces detected by theface detection unit 17a (S43). Then, thecandidate selection unit 26 selects a candidate face based on the size of the face (largest face) selected by the largest face selection unit 25 ( S44 ).

然后，决定部18e的中心决定部19e根据成为由候选选择部26所选择的候选的脸(最大脸、以及具有属于以该最大脸为基准的规定范围内的大小的至少一个其它脸)，取得脸点的中心(S45)。然后，决定部18e的脸决定部20e根据所取得的中心，选择一个脸，即成为对焦对象的脸(S46)。另外，根据上述处理，在未检测出与其它脸相当的脸的情况下，把最大脸决定为对象脸。Then, the center determination unit 19e of thedetermination unit 18e obtains a face (the largest face and at least one other face having a size within a predetermined range based on the largest face) that becomes a candidate selected by thecandidate selection unit 26. The center of the face point (S45). Then, the face determination unit 20e of thedetermination unit 18e selects one face, that is, a face to be focused on based on the acquired center (S46). In addition, according to the processing described above, when no face corresponding to another face is detected, the largest face is determined as the target face.

〔作用/效果〕〔Effect〕

根据第六实施方式中的摄像装置12e，仅根据具有某种程度的大小的脸，选择成为对焦对象的脸。某种程度的大小是根据最大脸的大小来判断的。According to theimaging device 12e in the sixth embodiment, a face to be focused is selected based on only faces having a certain size. A certain size is judged by the size of the largest face.

因此，可防止作为背景而被拍摄的人，即用户没有有意作为被摄物体的人脸被包含进来作为处理对象。因此，提高了选择包含成为对焦对象的脸的集合的处理、取得基于集合内所包含的脸的中心的处理、选择成为对焦对象的脸的处理的精度。Therefore, it is possible to prevent a person photographed as a background, that is, a human face that the user does not intend to be a subject, from being included as a processing object. Therefore, the accuracy of the process of selecting a set including a face to be focused, the process of obtaining a center based on the face included in the set, and the process of selecting a face to be focused are improved.

〔变形例〕〔Modification〕

最大脸选择部25也可以构成为根据所检测出的脸的肤色的区域的大小来进行选择。The largestface selection unit 25 may be configured to select based on the size of the detected face's skin color region.

并且，第二～第六实施方式也可以构成为具有第一实施方式中的脸检测部5来取代脸检测部17a～17e。Furthermore, the second to sixth embodiments may be configured to include theface detection unit 5 in the first embodiment instead of theface detection units 17a to 17e.

对象决定装置可被装配在进行基于最适于被摄物体的焦距或曝光等的摄影的摄像机和相机上。The subject determination device may be mounted on a video camera or camera that performs photography based on the focal length, exposure, etc. that are most suitable for the subject.

Claims (13)

1. an object determination device that is used for camera is characterized in that, comprising:

Detecting unit, it detects people's face from the image of being imported; And

The decision unit, when being detected a plurality of people's face by described detecting unit, it is according to the position of this a plurality of people's faces, and decision becomes the face of the object of focusing and/or becomes the face of the object controlled of exposing when implementing shooting from these a plurality of people's faces,

Described decision unit comprises:

Decision unit, center, it determines the center of the position of described a plurality of people's faces according to the position of described a plurality of people's faces; And

Face decision unit, it is according to the face of the described object of described center decision becoming.

2. an object determination device that is used for camera is characterized in that, comprising:

Frame is obtained the unit, and it obtains the image of a certain frame as the process object image from the sequential chart picture of being imported that comprises a plurality of frames;

Detecting unit, it detects people's face from described process object image; And

The decision unit, when being detected a plurality of people's face by described detecting unit, it is according to the position of this a plurality of people's faces, and decision becomes the face of the object of focusing and/or becomes the face of the object controlled of exposing when implementing shooting from these a plurality of people's faces,

Described decision unit comprises:

Decision unit, center, it determines the center of the position of described a plurality of people's faces according to the position of described a plurality of people's faces; And

Face decision unit, it is according to the face of the described object of described center decision becoming.

3. object determination device according to claim 1 and 2, wherein, described center determines the external polygonal center decision in the unit handle and the position of described a plurality of people's faces to be described center.

4. object determination device according to claim 1 and 2, wherein, decision unit, described center is described center to the decision of the center of gravity of the position of described a plurality of people's faces.

5. object determination device according to claim 1 and 2, wherein, described face decision unit is the people face decision nearest apart from described center the face that becomes described object.

6. object determination device according to claim 1 and 2, wherein, described face decision unit is benchmark according to the rules, from being arranged in apart from the face of the described object of face decision becoming of described center gauge set a distance.

7. object determination device according to claim 1 and 2, wherein, described decision unit the face decision that is positioned at below in described a plurality of people's faces for becoming the face of described object.

8. an object determination device that is used for camera is characterized in that, comprising:

Detecting unit, it detects people's face from the image of being imported; And

The decision unit, when detecting a plurality of people's face by described detecting unit, its number with detected face is a benchmark, is the people's face decision that is positioned at its center the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting.

9. an object determination device that is used for camera is characterized in that, comprising:

Frame is obtained the unit, and it obtains the image of a certain frame as the process object image from the sequential chart picture of being imported that comprises a plurality of frames;

Detecting unit, it detects people's face from described process object image; And

The decision unit, when detecting a plurality of people's face by described detecting unit, its number with detected face is a benchmark, is the people's face decision that is positioned at its center the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting.

10. according to any described object determination device in the claim 1,2, also comprise:

Taxon, when being detected a plurality of people's face by described detecting unit, it is categorized as a plurality of set with detected a plurality of people's faces according to separately position; And

Set decision unit, they are from described a plurality of set, and set is selected in decision, and this choice set share the face that becomes described object in decision,

Described decision unit is according to the people's face that is comprised in the described selection set, and decision becomes the face of described object.

11. any described object determination device according in the claim 1,2,8,9 is characterized in that, also comprises display unit, it distinguishes the demonstration of coming by people's face and other faces that described decision unit is determined.

12. an object determination device that is used for camera is characterized in that, comprising:

Face detecting unit, the image that any focus information of its basis is made a video recording be as the process object image, from described process object image, detects people's face according to the relative value because of the statistic of the concavo-convex of people's face or a plurality of characteristic areas that organ produces; And

The decision unit, it determines subject of focusing and/or the subject that becomes the object of the control that exposes according to by the detected face of described face detecting unit when the enforcement shooting,

When detecting a plurality of people's face by described face detecting unit, described decision unit is a benchmark with the number of detected face, is the people's face decision that is positioned at its center the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting.

13. an object determination device that is used for camera is characterized in that, comprising:

Frame is obtained the unit, and it is from the sequential chart picture that comprises a plurality of frames of making a video recording according to any focus information, and the image of obtaining a certain frame is as the process object image;

The face detecting unit, it detects people's face according to the relative value because of the statistic of the concavo-convex of people's face or a plurality of characteristic areas that organ produces from described process object image; And

The decision unit, it determines subject of focusing and/or the subject that becomes the object of the control that exposes according to by the detected face of described face detecting unit when the enforcement shooting,

When detecting a plurality of people's face by described face detecting unit, described decision unit is a benchmark with the number of detected face, is the people's face decision that is positioned at its center the face that becomes the face of the object of focusing and/or become the object of the control that exposes when implementing shooting.

Images