CN113961255A - Method and device for awakening face recognition - Google Patents

Abstract

The invention provides a method and a device for awakening face recognition, which are applied to an access control system provided with a human body sensing device and a camera, and the method comprises the following steps: if the human body sensing device senses that a person approaches, the camera is controlled to be started for shooting; if the face images meeting the living body detection requirements of a preset number of frames are continuously shot, the shot face images are uploaded to a cloud end to be subjected to face comparison; the cloud end stores face data of a white list; if the comparison result is that the door access control is passed, controlling the door access control to be opened; and if the face images meeting the living body detection requirement of the preset number of frames are not continuously shot, determining that the person passes by the face images unintentionally. The threshold for awakening the face recognition is improved by continuously capturing the face images of which the preset number of frames meet the living body detection requirement, and the problem of frequent awakening of the face recognition caused by the fact that people pass by or stay can be effectively solved.

Description

Method and device for awakening face recognition

Technical Field

The invention relates to the technical field of access control management, in particular to a method and a device for awakening face recognition.

Background

An existing face access control is usually internally provided with an infrared human body approaching sensor, and when a person passes through or approaches, face recognition is awakened; however, for the common passers-by, the short stop or the passing before the face access control can falsely trigger the face recognition and frequently prompt the voice to recognize the face and fail. Brings a feeling that the face information is collected all the time to passers-by, and is very unfriendly. Meanwhile, the collection of the part of non-frontal face images has no practical significance to machine or AI learning per se, and is a waste of resources.

Thus, there is a need for a better solution to the problems of the prior art.

Disclosure of Invention

The invention provides a method and a device for awakening face recognition, which can solve the technical problems that in the prior art, the face recognition is triggered by mistake when the face door is stopped for a short time or passes through the face door control, and the face recognition cannot pass through the face recognition by frequent voice prompt.

The technical scheme for solving the technical problems is as follows:

the embodiment of the invention provides a method for awakening face recognition, which is applied to an access control system provided with a human body sensing device and a camera, and comprises the following steps:

if the human body sensing device senses that a person approaches, the camera is controlled to be started for shooting;

if the face images meeting the living body detection requirements of a preset number of frames are continuously shot, the shot face images are uploaded to a cloud end to be subjected to face comparison; the cloud end stores face data of a white list;

if the comparison result is that the door access control is passed, controlling the door access control to be opened;

and if the face images meeting the living body detection requirement of the preset number of frames are not continuously shot, determining that the person passes by the face images unintentionally.

In a specific embodiment, the camera is a binocular camera.

In a specific embodiment, the human body sensing device is an infrared sensing device.

In a specific embodiment, the predetermined number is in the range of 12-20.

In a specific embodiment, the method further comprises:

and if the comparison result is that the vehicle passes the right, prompting that no passing authority exists.

The embodiment of the invention also provides a device for awakening face recognition, which is applied to an access control system provided with a human body sensing device and a camera, and comprises:

the starting module is used for controlling the camera to be started for shooting if the human body sensing device senses that a person approaches;

the comparison module is used for uploading the shot face images to a cloud end for face comparison if the preset number of frames of face images meeting the living body detection requirement are shot continuously; the cloud end stores face data of a white list;

the control module is used for controlling the entrance guard to open if the comparison result is that the entrance guard passes;

and the determining module is used for determining that a person passes by the living body detection module unintentionally if the preset number of frames of face images meeting the living body detection requirement are not continuously shot.

In a specific embodiment, the camera is a binocular camera.

In a specific embodiment, the human body sensing device is an infrared sensing device.

In a specific embodiment, the predetermined number is in the range of 12-20.

In a specific embodiment, the method further comprises the following steps:

and the prompting module is used for prompting that no passing permission exists if the comparison result is that the vehicle passes the pass permission.

The invention has the beneficial effects that:

the embodiment of the invention provides a method and a device for awakening face recognition, which are applied to an access control system provided with a human body sensing device and a camera, and the method comprises the following steps: if the human body sensing device senses that a person approaches, the camera is controlled to be started for shooting; if the face images meeting the living body detection requirements of a preset number of frames are continuously shot, the shot face images are uploaded to a cloud end to be subjected to face comparison; the cloud end stores face data of a white list; if the comparison result is that the door access control is passed, controlling the door access control to be opened; and if the face images meeting the living body detection requirement of the preset number of frames are not continuously shot, determining that the person passes by the face images unintentionally. The threshold for awakening the face recognition is improved by continuously capturing the face images of which the preset number of frames meet the living body detection requirement, and the problem of frequent awakening of the face recognition caused by the fact that people pass by or stay can be effectively solved.

Drawings

Fig. 1 is a schematic flow chart of a method for waking up face recognition according to an embodiment of the present invention;

fig. 2 is a schematic structural diagram of a device for waking up face recognition according to an embodiment of the present invention;

fig. 3 is a schematic structural diagram of a terminal according to an embodiment of the present invention;

fig. 4 is a schematic diagram of a frame structure of a storage medium according to an embodiment of the present invention.

Detailed Description

The principles and features of this invention are described below in conjunction with the following drawings, which are set forth by way of illustration only and are not intended to limit the scope of the invention.

Example 1

The method for awakening face recognition provided by the embodiment 1 of the invention is applied to an access control system provided with a human body sensing device and a camera, and as shown in fig. 1, the method comprises the following steps:

step S101, if a human body induction device induces that a person approaches, controlling to start the camera to shoot;

therefore, when the human body sensing device in the access control system detects that a person approaches, for example, the person approaches to a certain range (for example, within a range of 2 m), a control command is generated, and the camera is controlled to start to shoot based on the control command.

The specific camera and the human body sensing device are both arranged in an area door which is in a certain range in the access control system, for example, the distance between the camera and the human body sensing device is 0.5-4 m, namely the distance between the camera and the human body sensing device is 0.5-4 m, so that the simultaneous locality of data among the cameras is guaranteed.

In addition, in one embodiment, the camera is a binocular camera.

Specifically, the binocular camera is adopted, so that better image quality shooting can be realized, more accurate acquisition of the face image is facilitated, and the normal operation of the whole access control system is facilitated.

In addition, the human body sensing device is an infrared sensing device.

Specifically, when a person enters the sensing range, the infrared sensing device detects the change of the infrared spectrum of the human body, the load is automatically switched on, and the person is continuously switched on without leaving the sensing range; after the person leaves, the load is automatically closed in a delayed mode. Based on this, it is possible to detect whether or not a person approaches.

Step S102, if the human face images meeting the living body detection requirement are continuously shot in a preset number of frames, the shot human face images are uploaded to a cloud end to carry out human face comparison; the cloud end stores face data of a white list;

in a specific embodiment, the predetermined number is in the range of 12-20.

In one embodiment, for example, if 15 frames of face images satisfying the living body detection requirement are captured continuously (taking about 0.5 second), the face images are uploaded to the cloud permission library for passing permission verification (i.e., performing the face comparison process).

S103, if the comparison result is that the door access control is opened, controlling the door access control to be opened;

further, if the comparison result is that the user passes, that is, the shot face image is a face image of a white list, it indicates that the user is a white list user, and then the access control is opened.

Of course, if the comparison result is that the vehicle passes the right of passage, no permission is prompted.

And step S104, if the human face images meeting the living body detection requirement and continuously shooting a preset number of frames are not shot, determining that a person passes by the human face images unintentionally.

After step S101, if the preset number of frames of face images meeting the living body detection requirement are not continuously captured, it is determined that a person has passed by unintentionally, that is, the process of face comparison is not performed.

In a specific embodiment, the method comprises the following steps:

s1: face entrance guard's infrared induction device discerns someone and is close to

S2: starting to capture images by binocular camera

S3: if 15 frames of face images meeting the living body detection requirement are continuously captured (about 0.5 second is consumed), uploading the face images to a cloud permission base for pass permission verification, and executing S4

S4: and executing a face comparison process according to the face characteristic data, if the face is in the passing white list, opening the entrance guard, and if the face does not pass the passing white list, prompting no passing authority.

S5: if 15 frames of face images satisfying the living body requirement are not continuously captured, it is considered that a person has not passed through and S6 is executed.

S6: and if the person is judged to pass through the face comparison module unintentionally, the face comparison process is not executed.

Therefore, the scheme improves the threshold for awakening the face recognition by continuously capturing 15 frames of face images meeting the living body detection requirement, and can effectively solve the problem that the face recognition is frequently awakened due to the fact that people pass by or stay by accident.

Example 2

The embodiment 2 of the present invention further discloses a device for waking up face recognition, which is characterized in that the device is applied to an access control system provided with a human body sensing device and a camera, as shown in fig. 2, the device comprises:

the startingmodule 201 is used for controlling the camera to be started for shooting if the human body sensing device senses that a person approaches;

thecomparison module 202 is configured to upload the captured face images to a cloud for face comparison if a preset number of frames of face images meeting the living body detection requirement are continuously captured; the cloud end stores face data of a white list;

thecontrol module 203 is used for controlling the entrance guard to open if the comparison result is that the comparison result is passed;

the determiningmodule 204 is configured to determine that a person has passed through the living body detection module unintentionally if the preset number of frames of face images meeting the living body detection requirement are not continuously captured.

In a specific embodiment, the camera is a binocular camera.

In a specific embodiment, the human body sensing device is an infrared sensing device.

In a specific embodiment, the predetermined number is in the range of 12-20.

In a specific embodiment, the method further comprises the following steps:

and the prompting module is used for prompting that no passing permission exists if the comparison result is that the vehicle passes the pass permission.

Example 3

Embodiment 3 of the present invention further discloses a terminal, as shown in fig. 3, including a processor and a memory, where the memory stores a program, and when the processor runs the program in the memory, the processor executes the voice recognition method described in embodiment 1.

Example 4

Embodiment 4 of the present invention also discloses a storage medium, as shown in fig. 4, where an application program for executing the speech recognition method described in embodiment 1 is stored in the storage medium.

The embodiment of the invention provides a method and a device for awakening face recognition, which are applied to an access control system provided with a human body sensing device and a camera, and the method comprises the following steps: if the human body sensing device senses that a person approaches, the camera is controlled to be started for shooting; if the face images meeting the living body detection requirements of a preset number of frames are continuously shot, the shot face images are uploaded to a cloud end to be subjected to face comparison; the cloud end stores face data of a white list; if the comparison result is that the door access control is passed, controlling the door access control to be opened; and if the face images meeting the living body detection requirement of the preset number of frames are not continuously shot, determining that the person passes by the face images unintentionally. The threshold for awakening the face recognition is improved by continuously capturing the face images of which the preset number of frames meet the living body detection requirement, and the problem of frequent awakening of the face recognition caused by the fact that people pass by or stay can be effectively solved.

While the invention has been described with reference to specific embodiments, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (10)

Translated fromChinese

1.一种唤醒人脸识别的方法，其特征在于，应用于设置有人体感应装置与摄像头的门禁系统中，该方法包括：1. a method of awakening face recognition, is characterized in that, be applied in the access control system that is provided with human body induction device and camera, this method comprises:若人体感应装置感应到有人靠近，则控制启动所述摄像头进行拍摄；If the human body sensing device senses that someone is approaching, control to start the camera to shoot;若连续拍摄到预设数量帧的满足活体检测要求的人脸图像，则将拍摄到的人脸图像上传云端进行人脸比对；所述云端存储有白名单的人脸数据；If a preset number of frames of face images that meet the requirements of living body detection are continuously captured, upload the captured face images to the cloud for face comparison; the cloud stores the whitelisted face data;若比对结果为通过，则控制门禁开启；If the comparison result is passed, control the access control to open;若未连续拍摄到预设数量帧的满足活体检测要求的人脸图像，则确定为有人无意经过。If the preset number of frames of face images that meet the requirements of living body detection are not continuously captured, it is determined that someone has passed by unintentionally.2.如权利要求1所述的方法，其特征在于，所述摄像头为双目摄像头。2. The method of claim 1, wherein the camera is a binocular camera.3.如权利要求1所述的方法，其特征在于，所述人体感应装置为红外感应装置。3. The method of claim 1, wherein the human body sensing device is an infrared sensing device.4.如权利要求1所述的方法，其特征在于，所述预设数量的范围为12-20。4. The method of claim 1, wherein the preset number ranges from 12 to 20.5.如权利要求1所述的方法，其特征在于，还包括：5. The method of claim 1, further comprising:若比对结果为未通过，则提示无通行权限。If the comparison result is not passed, it will prompt no access permission.6.一种唤醒人脸识别的装置，其特征在于，应用于设置有人体感应装置与摄像头的门禁系统中，该装置包括：6. A device for wake-up face recognition, characterized in that it is applied in an access control system provided with a human body sensing device and a camera, the device comprising:启动模块，用于若人体感应装置感应到有人靠近，则控制启动所述摄像头进行拍摄；a start-up module for controlling and starting the camera to shoot if the human body sensing device senses that someone is approaching;比对模块，用于若连续拍摄到预设数量帧的满足活体检测要求的人脸图像，则将拍摄到的人脸图像上传云端进行人脸比对；所述云端存储有白名单的人脸数据；The comparison module is used for uploading the captured face images to the cloud for face comparison if a preset number of frames of face images that meet the requirements of living body detection are continuously captured; the cloud stores the faces of the white list data;控制模块，用于若比对结果为通过，则控制门禁开启；The control module is used to control the access control to open if the comparison result is passed;确定模块，用于若未连续拍摄到预设数量帧的满足活体检测要求的人脸图像，则确定为有人无意经过。The determining module is configured to determine that someone has passed by unintentionally if a preset number of frames of face images that meet the requirements of living body detection are not continuously captured.7.如权利要求6所述的装置，其特征在于，所述摄像头为双目摄像头。7. The apparatus of claim 6, wherein the camera is a binocular camera.8.如权利要求6所述的装置，其特征在于，所述人体感应装置为红外感应装置。8. The device of claim 6, wherein the human body sensing device is an infrared sensing device.9.如权利要求6所述的装置，其特征在于，所述预设数量的范围为12-20。9 . The device of claim 6 , wherein the preset number ranges from 12 to 20. 10 .10.如权利要求1所述的方法，其特征在于，还包括：10. The method of claim 1, further comprising:提示模块，用于若比对结果为未通过，则提示无通行权限。The prompt module is used to prompt that there is no access permission if the comparison result is failed.

Images