Device for collecting high-definition face imageTechnical Field
The invention relates to the field of biological feature acquisition and identification, in particular to a device for acquiring a high-definition facial image.
Background
In recent years, various biometric identification technologies have been widely applied in the fields of security protection, banks and the like, especially the market share of two technologies of face identification and iris identification has been steadily improved, and the trend of pursuing fingerprint identification technology is strong. However, the face recognition technology cannot be used for a large-scale database due to the recognition accuracy; the iris recognition technology is subject to the problem of user usability, mainly the problem that the user has poor interactive experience and cannot be applied in a large scale.
The face recognition and the iris recognition are combined, so that the perfect combination of the usability and the safety is achieved, and the popularization of the non-contact biological feature recognition technology is deeply influenced.
The optical module used for general face image acquisition has a large view field and a large acquisition range, but has low image resolution and low identification precision.
The general optical module for collecting the iris image has the advantages of small field angle, small collecting range, large image amplification rate, capability of shooting detailed texture information of an eye area and high identification precision.
The identification algorithm generally judges the image quality by using video stream, and if the acquired image is a high-definition image, the requirement on the bandwidth of information transmission is higher, and the requirement on the computing resource at the rear end is also stricter.
Disclosure of Invention
In view of the above problems, the present invention provides a device for collecting high-definition face images, which can collect high-definition face images and iris images at the same time, and provide basic data for a back-end recognition algorithm.
The technical scheme of the invention is as follows:
a device for collecting high-definition face images comprises a first intelligent camera, an illumination unit, a distance sensor and a holder, wherein the first intelligent camera comprises a face optical unit, a first iris optical unit, an algorithm module and a control module,
the distance sensor is used for measuring the object distance from a user to the first intelligent camera;
the first iris optical unit carries out automatic focusing according to the object distance;
the first intelligent camera and the lighting unit are fixed on the holder;
the holder comprises at least two motors, supports the rotation in the horizontal direction and the rotation in the vertical direction, and is used for adjusting the angles of the first intelligent camera and the lighting unit;
the face optical unit and the first iris optical unit are respectively connected with the first intelligent camera through MIPI interfaces;
the algorithm module carries out face detection on the collected face image, and adjusts the rotation angle of the holder according to the face detection result so that the face is positioned in the center of the image;
the algorithm module judges the image quality of the collected face image and transmits the face image meeting the quality requirement to the back end for processing;
the first intelligent camera is provided with a control module for logically controlling other peripheral devices contained in the device, the peripheral devices are optional devices and are added according to actual requirements, and when the first intelligent camera cannot meet the requirements of identifying distance and angle, the peripheral devices are added.
Preferably, the human face optical unit comprises a sensor with resolution not lower than 2M, an optical lens with FOV not less than 65 °, and MTF value of the human face optical unit in the collection range is not lower than 50% @0.5 lp/mm.
Preferably, the first iris optical unit comprises a sensor and a lens, the resolution ratio of the sensor is not lower than 8M, the MTF value of the first iris optical unit in an acquisition range is not lower than 50% @3lp/mm, and the spatial resolution ratio of pixels is greater than 16 pixel/mm.
Preferably, the first iris optical unit may employ a fixed focus structure or an auto focus motor.
Preferably, the lighting unit comprises one or more light-emitting components, the light-emitting components mix multiple spectra, and the light intensity and the angle are adjusted according to the distance information.
Preferably, the light emitting component is configured as an LED or VCSEL laser lighting module.
Preferably, the distance sensor is a TOF laser ranging sensor, a 3D structured light ranging sensor, an infrared ranging sensor, an ultrasonic ranging sensor or a lidar distance sensor.
Preferably, the peripheral device includes a second smart camera, and the second smart camera includes a second iris optical unit, is controlled by the control module of the first smart camera, and performs automatic focusing according to an object distance in cooperation with the first smart camera.
Preferably, the second intelligent camera is fixed on the holder and is connected with the second iris optical unit through an MIPI interface.
Preferably, the second iris optical unit comprises a sensor with resolution not lower than 8M and a lens, the MTF value of the second iris optical unit in an acquisition range is not lower than 50% @3lp/mm, and the spatial resolution of the pixel is greater than 16 pixel/mm.
Preferably, the second iris optical unit adopts a fixed focus structure or an automatic focusing motor.
Preferably, the peripheral device can also be configured with more intelligent cameras according to the requirements of actual application scenes.
Preferably, a method of acquiring a high-definition face image,
1) the method comprises the steps that a face optical unit is used for collecting a face image, an algorithm module detects the position and the size of a face through a face detection algorithm, the angle of a holder is adjusted according to the face position, and the face is adjusted to the center position of the whole image;
2) selecting to trigger the first iris optical unit or the second iris optical unit according to the size of the face or the object distance measured by the distance sensor, adjusting the focal length and simultaneously controlling the light intensity of the illumination unit;
3) the algorithm module judges the quality of the iris image through a quality judgment algorithm, selects the iris image meeting the quality requirement, and transmits the face image corresponding to the iris image to the rear end for image processing.
The invention has the beneficial effects that: according to the device for acquiring the high-definition facial image, the facial image is captured through the facial optical unit, the processing capacity of the first intelligent camera is enhanced, the high-definition facial image can be acquired, and the usability and accuracy of biological identification are met; only the high-definition face image is transmitted, so that the bandwidth requirement of data transmission is effectively reduced; the device has flexible structure and convenient installation and carrying, and can be additionally provided with an optical unit and an intelligent camera unit according to the requirement.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
Drawings
Further objects, features and advantages of the present invention will become apparent from the following description of embodiments of the invention, with reference to the accompanying drawings, in which:
FIG. 1 schematically illustrates a schematic diagram of an apparatus for acquiring high definition facial images in accordance with the present invention;
fig. 2 schematically shows a flow chart of the present invention for acquiring a high-definition face image using the apparatus for acquiring a high-definition face image.
Detailed Description
The objects and functions of the present invention and methods for accomplishing the same will be apparent by reference to the exemplary embodiments. However, the present invention is not limited to the exemplary embodiments disclosed below; it can be implemented in different forms. The nature of the description is merely to assist those skilled in the relevant art in a comprehensive understanding of the specific details of the invention.
Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the drawings, the same reference numerals denote the same or similar parts, or the same or similar steps.
Fig. 1 is a schematic diagram of an apparatus for acquiring a high-definition facial image according to the present invention, and as shown in fig. 1, the apparatus for acquiring a high-definition facial image according to the present embodiment includes a firstsmart camera 101, a distance sensor 103, a pan/tilt head 104, an illumination unit 105, a faceoptical unit 106, a first iris optical unit 107, an algorithm module (not shown), and a control module (not shown);
the embodiment further includes a secondsmart camera 102, where the secondsmart camera 102 is an optional device, and when the first smart camera cannot meet the requirements of expanding the recognition distance and expanding the recognition angle, the secondsmart camera 102 is added, and the secondsmart camera 102 is correspondingly matched with the second iris optical unit 108, where the secondsmart camera 102 is used as a slave smart camera of the device, and coordinates with the first smart camera according to a certain logic to extend the range of iris acquisition.
The distance sensor 103 is used for measuring an object distance from a user to the first smart camera, and may be a TOF laser ranging sensor, a 3D structured light ranging sensor, an infrared ranging sensor, an ultrasonic ranging sensor, or a laser radar distance sensor.
The first iris optical unit 107 and the second iris optical unit 108 perform automatic focusing according to the object distance.
The firstsmart camera 101, the secondsmart camera 102 and the lighting unit 105 are fixed on the pan-tilt 104, and the pan-tilt 104 includes at least two motors supporting rotation in the horizontal direction and rotation in the vertical direction for adjusting the angles of the firstsmart camera 101, the secondsmart camera 102 and the lighting unit 105.
The lighting unit 105 includes one or more light emitting devices that mix multiple spectra and enable light intensity adjustment based on distance information provided by the distance sensor 103 to the firstsmart camera 101.
The light-emitting component comprises an LED and a VCSEL laser lighting module.
When the faceoptical unit 106 captures a face and the firstsmart camera 101 collects a facial image of the user according to the captured face, the apparatus of the embodiment adjusts the holder 104 so that the face is finally located in the center of the whole image.
The faceoptical unit 106 and the first iris optical unit 107 are respectively connected with the firstsmart camera 101 through MIPI interfaces, and the second iris optical unit 108 is respectively connected with the secondsmart camera 102 through MIPI interfaces.
The human faceoptical unit 106 comprises a sensor with resolution not lower than 2M, an optical lens with FOV not less than 65 degrees, and the MTF value of the human face optical unit in the collection range is not lower than 50% @0.5 lp/mm.
The first iris optical unit 107 and the second iris optical unit 108 comprise a sensor with resolution not lower than 8M and a lens, MTF values of the first iris optical unit 107 and the second iris optical unit 108 in an acquisition range are not lower than 50% @3lp/mm, and spatial resolution of pixels is larger than 16 pixel/mm.
The first iris optical unit 107 and the second iris optical unit 108 may adopt a fixed focus structure or an auto-focus motor.
The firstintelligent camera 101 serves as a main intelligent camera, an algorithm module is arranged in the first intelligent camera, the first intelligent camera has image processing capacity, the quality judgment can be carried out on the acquired iris image, a threshold value is set, and when the first intelligent camera detects that the quality of the acquired image reaches the preset threshold value, the first intelligent camera transmits a face image corresponding to the first intelligent camera to a background to carry out iris registration and iris recognition; otherwise, automatically discarding the iris and face images acquired at this time, and continuing to acquire the images.
The firstsmart camera 101 includes a control module that enables logical control of the secondsmart camera 102.
Fig. 2 schematically shows a flow chart of the present invention for acquiring a high-definition facial image by using the apparatus, as shown in fig. 2, specifically including the following steps:
s1: starting an acquisition process;
s2: a face optical unit of a first intelligent camera collects a face image;
s3: judging whether the face is in a target area; if yes, go to step S5, otherwise, go to step S4;
s4: adjusting the position of the holder;
the first intelligent camera detects the position and the size of a face from the image, controls the holder according to the face position and adjusts the face to be close to the center position of the whole image;
s5: reading distance information;
the distance sensor provides the first smart camera with an object distance of the user to the first smart camera;
s6: selecting an iris optical unit and adjusting a focal length;
determining to use the first iris optical unit or the second iris optical unit according to the size of the face of the user or the object distance measured by the distance sensor, adjusting the focal length to enable the first intelligent camera to be aligned to the eyebrow area of the face, and continuously tracking; meanwhile, the light intensity of the lighting unit is controlled, and relatively similar lighting conditions are ensured to be obtained at different distances;
s7: reading an iris image;
s8: judging whether the iris image meets the quality requirement, if so, performing step S9, otherwise, returning to step S5, and reading the distance information again;
the first intelligent camera setting algorithm module is used for judging the quality of the acquired image, setting a threshold value and judging whether the acquired image meets the quality requirement;
s9: pushing a high definition face image;
the first intelligent camera continuously performs automatic focusing and image quality judgment, extracts clear images (human faces and eye marks) and pushes the clear images to the rear-end image processing platform;
s10: the collection process is ended.
According to the device for acquiring the high-definition facial image, data acquisition is performed at the front end of the acquisition equipment, and the image judged by the quality through the main intelligent camera is transmitted to the background for processing, so that the pressure of transmission bandwidth is reduced; the invention can acquire the high-definition face image and simultaneously meet the usability and accuracy of biological identification.
Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.