CN109447597B

CN109447597B - Method and device for attendance checking by multiple persons and face recognition system

Info

Publication number: CN109447597B
Application number: CN201811615540.5A
Authority: CN
Inventors: 刘毓森
Original assignee: 深圳市沃特沃德信息有限公司
Current assignee: Shenzhen Waterward Information Co Ltd
Priority date: 2018-12-27
Filing date: 2018-12-27
Publication date: 2022-04-12
Anticipated expiration: 2038-12-27
Also published as: CN109447597A

Abstract

The invention discloses a method and a device for checking attendance by multiple persons and a face recognition system, wherein the method for checking attendance by multiple persons is applied to the face recognition system, the face recognition system comprises a first camera and a second camera, and the method comprises the following steps: acquiring a current image of a designated area acquired by a first camera; analyzing whether one or more attendance checking persons exist in the designated area or not according to the current image; if one or more attendance checking persons exist in the designated area, analyzing the attendance checking persons to form an analysis result; determining whether to start a second camera according to the analysis result so as to acquire a first face image; and performing face recognition attendance according to the first face image. The method for attendance by multiple people can be used for the multiple people to carry out face recognition attendance, and the attendance efficiency based on face recognition is effectively improved.

Description

Method and device for attendance checking by multiple persons and face recognition system

Technical Field

The invention relates to the technical field of face recognition, in particular to a method and a device for attendance checking by multiple persons and a face recognition system.

Background

With the development of face recognition technology, face recognition technology has been widely applied to the fields of attendance checking, entrance guard, monitoring, security protection and the like.

At present, the face images of staff are generally collected through a fixed camera on a face recognition terminal in attendance based on face recognition, and then face recognition attendance is carried out according to the collected face images, but face recognition of one person can only be completed in the mode at each time, the attendance checking efficiency is low, and in order to improve the efficiency of face recognition attendance checking, a general solution is to solve by increasing the number of the face recognition terminals, but the use cost is undoubtedly increased.

Therefore, how to effectively improve attendance checking efficiency based on face recognition is a technical problem to be urgently solved by technical personnel in the field.

Disclosure of Invention

The invention mainly aims to provide a method and a device for attendance checking by multiple persons and a face recognition system.

The invention provides a method for checking attendance by multiple persons, which is applied to a face recognition system, wherein the face recognition system comprises a first camera and a second camera, and the method comprises the following steps:

acquiring a current image of a designated area acquired by a first camera;

analyzing whether one or more attendance checking persons exist in the designated area or not according to the current image;

if one or more attendance checking persons exist in the designated area, analyzing the attendance checking persons to form an analysis result;

determining whether to start a second camera according to the analysis result so as to acquire a first face image;

and performing face recognition attendance according to the first face image.

Further, the step of analyzing the attendance checking person to form an analysis result comprises the following steps:

matching a second face image of the attendance in the current image with a plurality of preset face images in the database, and analyzing whether an attendance with unsuccessful matching exists or not;

and if so, analyzing and matching the position information of the visual angle range of the second camera corresponding to the unsuccessful attendance checking person to form an analysis result.

Further, the step of determining whether to turn on the second camera according to the analysis result to acquire the first face image includes:

and if the analysis result comprises the position information of the visual angle range of the second camera corresponding to at least one attendance, starting the second camera to acquire the first face image.

Further, the step of analyzing whether one or more attendance checking persons exist in the specified area according to the current image comprises the following steps:

performing human head detection on the current image to obtain human head information in the current image, wherein the human head information comprises the number of human heads;

judging whether the number of the human heads is larger than or equal to a preset value;

and if the number of the heads is larger than or equal to the preset value, judging that one or more attendance checking persons exist in the designated area.

Further, first face image is a plurality of, carries out the step of face identification attendance according to first face image, includes:

respectively configuring a designated identifier for each first face image;

similarity matching is carried out on each first face image configured with the designated identifier and a plurality of preset face images in a database respectively, and a preset face image with the maximum similarity to the first face image is obtained, wherein the preset face image is associated with corresponding preset user information;

judging whether the similarity between the first face image and a preset face image is greater than a preset threshold value or not;

if yes, outputting corresponding first prompt information according to preset user information.

Further, after the step of determining whether the similarity between the first face image and the preset face image is greater than the preset threshold, the method further includes:

and if the similarity between the first face image and the preset face image is not greater than the preset threshold value, outputting corresponding second prompt information according to the designated identifier on the first face image.

Further, after the step of outputting the corresponding first prompt information according to the preset user information, the method further includes:

and replacing the preset face image matched with the first face image in the database with the first face image.

Further, after the step of performing face recognition attendance according to the first face image, the method further comprises the following steps of:

counting the number of times of face identification attendance failure in an attendance time period, and judging whether the number of times is greater than a preset number of times;

if yes, video information collected by the first camera in the attendance time period is saved.

The invention also provides a device for checking attendance by multiple persons, which is applied to a face recognition system, wherein the face recognition system comprises a first camera and a second camera, and the device comprises:

the acquisition module is used for acquiring a current image of the appointed area acquired by the first camera;

the first analysis module is used for analyzing whether one or more attendance checking persons exist in the specified area or not according to the current image;

the second analysis module is used for analyzing the attendance checking persons to form an analysis result when one or more attendance checking persons exist in the specified area;

the determining module is used for determining whether to start the second camera according to the analysis result so as to acquire a first face image;

and the face recognition attendance module is used for carrying out face recognition attendance according to the first face image.

The invention also provides a face recognition system, which comprises a first camera, a second camera, a memory, a processor and an application program, wherein the first camera and the second camera are respectively connected to the processor, the application program is stored in the memory and configured to be executed by the processor, and the application program is configured to be used for executing the method for attendance checking by multiple persons.

The invention has the beneficial effects that: the method for attendance checking by multiple persons in the embodiment of the invention utilizes the first camera to monitor the designated area, when one or more attendance persons exist in the designated area through the first camera, firstly, the first camera is used for carrying out face recognition attendance checking, then, for the attendance persons who fail to finish face recognition attendance checking preliminarily through the first camera, the corresponding second camera is opened according to the designated position of each attendance person, and each second camera is used for finishing corresponding face recognition attendance checking respectively, so that the attendance persons in the designated area can carry out face recognition attendance checking jointly, and the attendance checking efficiency based on face recognition is effectively improved.

Drawings

Fig. 1 is a schematic flow chart of a method for attendance checking by multiple persons in combination according to an embodiment of the present invention;

fig. 2 is a schematic structural diagram of an apparatus for attendance checking by multiple persons in accordance with an embodiment of the present invention;

FIG. 3 is a schematic diagram showing a detailed structure of the first analysis module in FIG. 2;

FIG. 4 is a schematic diagram showing a specific structure of the second analysis module in FIG. 2;

fig. 5 is a schematic structural diagram of the face recognition attendance module in fig. 2;

fig. 6 is a schematic structural diagram of an apparatus for attendance checking by multiple persons in accordance with another embodiment of the present invention;

FIG. 7 is a schematic diagram of the use of a face recognition system in an embodiment of the present invention;

FIG. 8 is a schematic plan view of a current image in an embodiment of the invention;

FIG. 9 is a schematic plan view of a comparison image in accordance with an embodiment of the present invention;

FIG. 10 is a schematic diagram illustrating a comparison between a current image and a comparison image according to an embodiment of the present invention.

The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.

Detailed Description

It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

Referring to fig. 1 and 7, the present invention provides a method for checking attendance by multiple persons, which is applied to a face recognition system, wherein the face recognition system includes a first camera and a plurality of second cameras, a view angle range of the first camera covers a designated area, view angle ranges of the second cameras respectively cover designated positions, and the designated positions are covered in the designated area, and the method includes the following steps:

s11, acquiring a current image of the first camera in the acquired appointed area;

s12, analyzing whether one or more attendance checking persons exist in the designated area according to the current image;

if yes, executing S13, and analyzing the attendance checking person to form an analysis result;

s14, determining whether to start the second camera according to the analysis result so as to acquire a first face image;

and S15, performing face recognition attendance according to the first face image.

In S11, the view angle range of the first camera covers the designated area, and the first camera captures the designated area to obtain current images of consecutive frames, so as to perform the related operations in the following step; and each second camera can be set according to a certain angle or/and set at a certain distance from each second camera, so that each camera can cover a designated position, for example, three second cameras (named A, B, C respectively) can be installed on the face recognition system, and by reasonably arranging the installation angle of each second camera and the distance between each second camera, the second camera a can be used for covering a designated position a, the second camera B can be used for covering a designated position B, and the second camera C can be used for covering a designated position C, wherein the designated positions a, B and C are all covered in a designated area, preferably, the designated positions a, B and C are mutually independent and do not interfere, and in the specific implementation, the designated position a can be divided into a plurality of positions in the ground area, b. c are separated from each other, for example, three circular or square areas can be drawn on the ground to represent the designated positions a, b and c respectively, and the three circular or square areas are separated from each other and do not interfere with each other, so that when face recognition attendance checking is performed, an attendance checker only needs to stand on one circular or square area.

In S12, since the current image includes the object information appearing in the designated area and the position information of the object in the designated area, it is possible to know whether or not the attendance checking person is present in the designated area and the position of the attendance checking person in the designated area by analyzing the current image, where "a plurality" means two or more.

In S13, specifically, when it is known that one or more attendance check users exist in the designated area through the current image, the face feature information corresponding to each attendance check user in the current image is analyzed, so as to form an analysis result, where the analysis result includes that each attendance check user completes face recognition attendance check, each attendance check user does not complete face recognition attendance check, and some attendance check users complete face recognition attendance check.

In the above S14, if the analysis result indicates that each attendance checker completes face recognition attendance, it indicates that the face recognition system completes identity verification of each attendance checker through the current image acquired by the first camera, and at this time, the face recognition system does not need to start the second camera; if the analysis result shows that each attendance checking person does not complete face recognition attendance checking or only part of attendance checking persons complete face recognition attendance checking, it is indicated that the face recognition system has a condition that the identity of the attendance checking person cannot be recognized through information in the current image, for example, face characteristic information cannot be detected in the current image or face characteristic information is detected, but part of the face characteristic information is incomplete, so that the identity of the attendance checking person cannot be recognized, and at the moment, the face recognition system starts a second camera to collect a first face image of the attendance checking person with unsuccessful primary attendance checking.

In the above-mentioned S15, the face recognition system analyzes the current image, and can know the designated position of each attendance checker (including an attendance checker with unsuccessful preliminary attendance checking) in the designated area, and then opens the corresponding second camera according to the designated position of the attendance checker with unsuccessful preliminary attendance checking, so as to acquire the corresponding first face image, and further realize attendance checking of the attendance checker by performing face recognition on the first face image.

The method for attendance checking by multiple persons in the embodiment of the invention utilizes the first camera to monitor the designated area, when one or more attendance persons exist in the designated area through the first camera, firstly, the first camera is used for carrying out face recognition attendance checking, then, for the attendance persons who fail to finish face recognition attendance checking preliminarily through the first camera, the corresponding second camera is opened according to the designated position of each attendance person, and each second camera is used for finishing corresponding face recognition attendance checking respectively, so that the attendance persons in the designated area can carry out face recognition attendance checking jointly, and the attendance checking efficiency based on face recognition is effectively improved.

In a preferred embodiment, before the step of acquiring the current image of the designated area acquired by the first camera, the method further includes:

s10a, detecting whether an attendance person approaches the face recognition system;

and if the attendance checking person approaches the face recognition system, executing S10b, and starting the first camera to acquire the current image of the designated area.

In this embodiment, specifically, the face recognition system may detect whether an attendee is close to the face recognition system through an existing human body sensing technology, for example, an infrared sensor or a proximity sensor may be arranged on the face recognition system to detect whether an attendee is close to the face recognition system, and when it is detected that an attendee is close to the face recognition system, the face recognition system starts the first camera according to a signal fed back by the infrared sensor or the proximity sensor to acquire a current image of the designated area, so that the attendance may be acquired when the attendance is close to the face recognition system, the first camera is prevented from being continuously started, and a power saving effect is achieved; of course, the effect of saving power can also be achieved by setting to turn on the first camera at regular time through the face recognition system, for example, for an enterprise facing nine nights and six nights, the time period for turning on the first camera can be set to be 7 o 'clock to 10 o' clock in the morning and five o 'clock in the afternoon to eight o' clock in the evening.

In a preferred embodiment, the step of analyzing the attendance checker to form an analysis result comprises:

s131, respectively matching second face images of the attendance in the current image with a plurality of preset face images in a database, and analyzing whether the attendance with unsuccessful matching exists or not, wherein the preset face images are associated with corresponding preset user information;

if so, executing S132, and analyzing the position information of the visual angle range of the second camera corresponding to the attendance checking person with unsuccessful matching to form an analysis result;

if not, executing S133, and outputting corresponding first prompt information according to preset user information.

In the above S131, specifically, the face recognition system performs face detection on the current image, may analyze a second face image existing in the current image, then matches the detected second face image with a plurality of preset face images in the database one by one, may obtain a corresponding matching score, and when a matching score corresponding to a certain second face image is greater than a preset threshold, it indicates that the identity verification of the attendance corresponding to the second face image is successful, that is, there is an attendance matching successfully; and when the matching score corresponding to a certain second face image is smaller than a preset threshold value, the authentication of the attendance corresponding to the second face image fails, namely, the attendance with unsuccessful matching exists.

In the above S132, for example, referring to fig. 7 to 10, a comparison image 3 may be stored in the face recognition system in advance, where the comparison image 3 is a two-dimensional planar image corresponding to the designated area preset in the face recognition system when no attendance exists in the designated area, the comparison image 3 includes a preset area, the preset area is a position of the designated position in the comparison image 3, that is, the position of the preset area in the comparison image 3 is equivalent to the position of the designated position in the designated area, for example, the designated positions a, b, and c exist in the designated area, so that three one-to-one corresponding preset areas one 31, two 32, and three 33 exist in the comparison image 3, and the preset area may be a circular area or a square area, and the specific shape may be preset by the face recognition system; when the face recognition system performs face detection (or head detection) on the current image 2, a face (or head) area 21 (i.e., an area where a face exists) in the current image 2 can be extracted by performing operations such as morphological processing, light compensation and the like on the current image 2, the face area 21 can be a circular area or a square (rectangular or square) area, and the specific shape can be preset by the face recognition system; specifically, whether the face area 21 is located within the preset area may be determined by comparing the current image 2 with the comparison image 3 and determining whether the center of the face area 21 falls within the preset area, for example, the face area 21 is a circular area, when the center of the circle of the circular area falls within the preset area, it indicates that the face area 21 falls within the preset area, and when the center of the circle of the circular area does not fall within the preset area, it indicates that the face area 21 does not fall within the preset area; when the face area 21 falls within the preset area, it is indicated that an attendance is standing at the designated position, and then the designated position of each attendance can be determined according to the preset area in which the face area 21 falls, for example, when two face areas 21 in the comparison image 3 obtained by comparing the current image 2 with the comparison image 3 respectively fall within the first preset area 31 and the second preset area 32, it is indicated that an attendance exists at the designated position a and the designated position b, and an attendance does not exist at the designated position c; therefore, the face recognition system can know whether attendance checking persons exist in the designated area or not and the positions of each attendance checking person (including attendance checking persons with unsuccessful primary attendance checking) in the designated area, and then the face recognition system can conveniently start a corresponding second camera to acquire a first face image subsequently.

In the above S133, specifically, before performing face recognition attendance, face images and user information of all attendance checking persons are pre-entered by a face recognition system, and a database is established, where the face images of the attendance checking persons correspond to the user information one by one (the preset face images are associated with corresponding preset user information); specifically, because the preset face images correspond to the preset user information one to one, after the attendance identity verification is successful, the face recognition system can recognize which attendance has finished the card punching according to the preset user information, and accordingly the face recognition system can output a word or voice of "a certain person, the card punching is successful", for example, to remind the corresponding attendance of finishing the card punching.

In a preferred embodiment, the step of determining whether to turn on the second camera to capture the first face image according to the analysis result includes:

and S14a, if the analysis result includes the position information of the visual angle range of the second camera corresponding to at least one attendance, starting the second camera to collect the first face image.

In this embodiment, the analysis result includes that each attendance has completed face recognition attendance, each attendance has not completed face recognition attendance, and some attendance has completed face recognition attendance, and specifically, when the analysis result is that each attendance has not completed face recognition attendance or some attendance has completed face recognition attendance, the corresponding analysis result further includes position information of a second camera view angle range corresponding to at least one attendance, so that the face recognition system subsequently opens a corresponding second camera according to the position information to acquire a first face image of an attendance with unsuccessful preliminary attendance.

In a preferred embodiment, the step of analyzing whether one or more attendance checking persons exist in the designated area according to the current image comprises the following steps:

s121, performing human head detection on the current image to obtain human head information in the current image, wherein the human head information comprises the number of human heads;

s122, judging whether the number of the heads is larger than or equal to a preset value;

if yes, S123 is executed to determine that one or more attendance checking users exist in the designated area.

In the above S121, the face recognition system receives the current image from the first camera and performs the human head detection on the current image by using the existing human head detection technology, so as to obtain the human head information in the current image, where the human head information includes the number of human heads, so as to perform the related operations in the following.

In S122, the face recognition system compares the number of the human heads in the obtained human head information with a preset value, and determines whether the number of the human heads in the current image is greater than or equal to the preset value, where the preset value is greater than or equal to 1.

In the above S123, if the number of the heads in the current image is greater than or equal to the preset value, it indicates that one or more attendance check users exist in the designated area, and therefore the face recognition system can determine that one or more attendance check users exist in the designated area accordingly.

In a preferred embodiment, the number of the first face images is multiple, and the step of performing face recognition attendance according to the first face images includes:

s151, respectively configuring a designated identifier for each first face image;

s152, performing similarity matching on each first face image configured with the designated identifier and a plurality of preset face images in a database respectively to obtain a preset face image with the maximum similarity to the first face image, wherein the preset face image is associated with corresponding preset user information;

s153, judging whether the similarity between the first face image and a preset face image is larger than a preset threshold value or not;

if yes, go to step S154, and output a corresponding first prompt message according to the preset user information.

If not, S155 is executed, and corresponding second prompt information is output according to the designated identifier on the first face image.

In S151 above, specifically, for example, after the face recognition system acquires the first face image captured by the second camera a, a specified identifier, for example, the number "1", may be configured for the first face image, and similarly, a specified identifier, for example, the number "2", may be configured for the first face image captured by the second camera B; a designated identifier, for example, the number "3", may be configured for the first face image captured by the second camera C, so that the first face image captured by each second camera may be distinguished by the designated identifier, that is, the identifier may be used to determine from which designated position the first face image is from the attendance checker, so as to perform subsequent operations.

In the above S152, specifically, before performing face recognition attendance, face images and user information of all attendance checking persons are pre-entered by a face recognition system, and a database is established, where the face images of the attendance checking persons correspond to the user information one by one (that is, the preset face images are associated with corresponding preset user information); during verification, the face recognition system respectively carries out similarity matching on the acquired first face images and a plurality of preset face images in a database one by one so as to search a corresponding matching image with the highest similarity from the database, for example, the face recognition system firstly obtains a first face image '1' from a second camera A, then the face recognition system carries out similarity matching on the first face image '1' and the plurality of preset face images in the database one by one so as to obtain a plurality of similarities, then screens out a preset face image with the highest similarity, the preset face image is the face image which is preliminarily matched with the first face image '1' in the database, and when the first face image '2' from the second camera B or the first face image '3' from the second camera C is obtained, the similarity matching process is similar to the process, this will not be described in detail.

In the above S153, when the matched preset face image is obtained, the similarity between the first face image and the preset face image is compared with a preset threshold, and whether the similarity is greater than the preset threshold is determined, if yes, it indicates that the first face image obtained by the face recognition system and the preset face image in the database are both from the same person, and the attendance identity verification is successful.

In the above S154, the first prompt information may be presented in a form of text, or in a form of voice, or in a form of a combination of text and voice, specifically, since the face image preset in the database by the attendance is in one-to-one correspondence with the user information, after the attendance is successfully authenticated, the face recognition system may recognize which attendance has completed the card punching according to the user information, and accordingly, the face recognition system may output, for example, a word or voice "a certain person, a successful card punching" to remind the corresponding attendance that the card punching has been completed.

In the above S155, specifically, if the similarity between the first face image and the preset face image is not greater than the preset threshold, it indicates that the first face image acquired by the face recognition system and the preset face image in the database are not from the same person, and the attendance checking authentication fails, at this time, since the first face images acquired by the face recognition system are all configured with the specified identifier, the face recognition system may recognize which attendance checking person at the specified position fails to be successfully checked according to the specified identifier, for example, when the first face image "1" acquired by the first camera a fails to be successfully matched, the face recognition system may determine that the first face image failed to be successfully matched is from the attendance checking person at the position a according to the specified identifier "1", so that the face recognition system may accordingly output, for example, "position a fails to be checked, please re-try the words or voices to remind the corresponding attendance to punch the card again.

In a preferred embodiment, after the step of outputting the corresponding first prompt information according to the preset user information, the method further includes:

s154a, acquiring the current attendance time;

s154b, judging whether the current attendance time exceeds a preset time interval;

if not, executing S154c, and generating corresponding first attendance information according to the current attendance time and preset user information;

if yes, executing S154d, and generating corresponding second attendance information according to the current attendance time and the preset user information.

In the above S154a, the preset user information includes personal information such as name, department, and position of the attendance, and after the attendance completes the identity verification, the face recognition system obtains the current attendance time of the attendance so as to perform the related operations in the following.

In the above S154b, the preset time interval may be determined according to actual conditions, for example, for a business facing nine nights and six nights, the preset time interval may be set to seven to nine am and six to eight pm.

In the above-mentioned S154c, if the current attendance time does not exceed the preset time interval, it indicates that the attendance person completed attendance within the specified time, and at this time, the first attendance information is normal, that is, the first attendance information is displayed with the evaluation result in addition to the personal information and the attendance time of the attendance person, and the evaluation result can be displayed through a word of "attendance normal", so that the administrator can quickly know the attendance result of the attendance person, and meanwhile, the administrator can conveniently count the attendance condition of the attendance person.

In the above-mentioned S154d, if the current attendance time exceeds the preset time interval, it indicates that the attendance person completed attendance within the unspecified time, at this time, the second attendance information is abnormal, that is, the second attendance information is displayed with an evaluation result in addition to the personal information and the attendance time of the attendance person, and the evaluation result can be displayed through a word of "attendance abnormal", so that the administrator can quickly know the attendance result of the attendance person, and the administrator can conveniently count the attendance condition of the attendance person.

In this embodiment, through setting up preset time interval, be favorable to supervising and urging staff to accomplish attendance in the appointed time to prevent that the staff from punching the card in advance too and the emergence of the phenomenon that the person is not on the scene, simultaneously through generating attendance information, the administrator of being convenient for is makeed statistics of and is evaluateed attendance condition to the attendance.

S154A, replacing the preset face image matching the first face image in the database with the first face image.

In this embodiment, when a certain first face image acquired by the face recognition system is successfully matched with a certain preset face image in the preset database, the preset face image is replaced with the first face image to update the preset face image corresponding to the attendance in the database, and as the time interval is longer, the difference between the acquired first face image and the preset face image in the database may be larger, so that normal attendance of the attendance may not be performed under the condition that the identity of the attendance is real, therefore, by dynamically updating the preset face image corresponding to the attendance in the database, the risk of face image verification failure due to overlong time interval is favorably reduced, and the stability of face recognition attendance is improved.

In a preferred embodiment, after the step of performing face recognition attendance according to the first face image, the method further includes:

s16, counting the frequency of face recognition attendance failure in the attendance time period, and judging whether the frequency is greater than the preset frequency;

if yes, S17 is executed, and video information collected by the first camera in the attendance time period is saved;

if not, executing S18, and not storing the video information collected by the first camera in the attendance time period.

In this embodiment, specifically, the first camera may adopt a monitoring camera having a real-time video recording function and a shooting function on the market, and in the specific implementation, the face recognition system may set the real-time video recording function of the first camera to be started at regular time to acquire video information of a designated area, for example, for an enterprise facing nine nights and six nights, the time period for the first camera to start the real-time video recording function may be set to be 7 o 'clock to 10 o' clock in the morning and five o 'clock to eight o' clock in the afternoon; meanwhile, whether an attendance person approaches the face recognition system or not can be detected by arranging an infrared sensor or a proximity sensor on the face recognition system, and when the attendance person is detected to approach the face recognition system, the face recognition system starts a shooting function of a first camera according to a signal fed back by the infrared sensor or the proximity sensor to acquire a current image of a designated area; specifically, after face recognition attendance is respectively performed according to the first face images acquired by the second cameras, the number of face recognition attendance failures in an attendance time period can be counted by the face recognition system, for example, the morning time period is 7 to 10 am, and then the number of face recognition attendance failures in the time period can be counted before 10 am, wherein the reason of the face recognition attendance failures may be due to cheating by an attendance (for example, cheating by a photo), or due to an excessively high setting of a preset threshold (for verification of the identity of the attendance and the consistency of the above); when the number of times of face recognition attendance failure is judged to be greater than the preset number of times, video information collected by the first camera in the early attendance time period is stored, wherein the preset number of times can be determined according to the actual conditions of enterprises, so that whether an attendance person cheats or not can be known by checking the video information, if the cheating phenomenon is not found through the video information, the reason that the preset threshold value is set too high is probably shown, and at the moment, a manager can appropriately adjust the preset threshold value built in the face recognition system; when the frequency of face recognition attendance failure is judged to be not more than the preset frequency, the video information collected by the first camera in the morning time period is not stored, so that the storage space of a face recognition system is saved; therefore, in the implementation, the purpose of face recognition attendance can be achieved, the attendance condition in the designated area can be monitored in the attendance time period through the first camera, so that the occurrence of the phenomenon that an attendance person replaces attendance can be reduced without increasing a living body detection technology, meanwhile, the preset threshold value can be practically adjusted in a matching manner, and the flexibility and the practicability of the face recognition system are improved.

Referring to fig. 2 and 7, an embodiment of the present invention further provides a device for attendance checking by multiple people, which is applied to a face recognition system, where the face recognition system includes a first camera and a second camera, and the device includes:

the acquisition module 11 is configured to acquire a current image of an appointed area acquired by a first camera;

the first analysis module 12 is configured to analyze whether one or more attendance checking persons exist in the designated area according to the current image;

the second analysis module 13 is configured to, when one or more attendance checking users exist in the designated area, analyze the attendance checking users to form an analysis result;

a determining module 14, configured to determine whether to turn on the second camera according to the analysis result to acquire the first facial image;

and the face recognition attendance module 15 is used for carrying out face recognition attendance according to the first face image.

Specifically, the view angle range of the first camera covers the designated area, and the acquisition module 11 captures the designated area through the first camera, so as to obtain current images of consecutive multiple frames, so as to execute the related operations in the following; and each second camera can be set according to a certain angle or/and set at a certain distance from each second camera, so that each camera can cover a designated position, for example, three second cameras (named A, B, C respectively) can be installed on the face recognition system, and by reasonably arranging the installation angle of each second camera and the distance between each second camera, the second camera a can be used for covering a designated position a, the second camera B can be used for covering a designated position B, and the second camera C can be used for covering a designated position C, wherein the designated positions a, B and C are all covered in a designated area, preferably, the designated positions a, B and C are mutually independent and do not interfere, and in the specific implementation, the designated position a can be divided into a plurality of positions in the ground area, b. c are separated from each other, for example, three circular or square areas can be drawn on the ground to represent the designated positions a, b and c respectively, and the three circular or square areas are separated from each other and do not interfere with each other, so that when face recognition attendance checking is performed, an attendance checker only needs to stand on one circular or square area.

Since the current image includes the object information appearing in the designated area and the position information of the object in the designated area, whether the attendance checker exists in the designated area and the position of the attendance checker in the designated area can be known by analyzing the current image with the first analysis module 12, where "a plurality" means two or more.

When the first analysis module 12 learns that one or more attendance checking persons exist in the designated area, the second analysis module 13 analyzes the face feature information corresponding to each attendance checking person in the current image respectively to form an analysis result, wherein the analysis result comprises that each attendance checking person completes face recognition attendance checking, each attendance checking person does not complete face recognition attendance checking, and a part of attendance checking persons complete face recognition attendance checking.

If the analysis result shows that each attendance checking person completes face recognition attendance checking, the face recognition system completes identity verification of each attendance checking person through the current image acquired by the first camera, and the determining module 14 can determine not to start the second camera according to the identity verification; if the analysis result shows that each attendance checking person does not complete face recognition attendance checking, or only part of attendance checking persons complete face recognition attendance checking, it indicates that the face recognition system has a condition that the attendance checking person identity cannot be recognized through information in the current image, for example, face feature information cannot be detected in the current image, or face feature information is detected, but part of face feature information is incomplete, so that the attendance checking person identity cannot be recognized, and at this time, the determining module 14 may determine to open the second camera according to the condition to acquire the first face image of the attendance checking person with unsuccessful primary attendance checking.

The face recognition system analyzes the current image through the first analysis module 12, can learn the designated position of each attendance (including the attendance unsuccessful in primary attendance) in the designated area, then open the corresponding second camera through the determination module 14 according to the designated position of the attendance unsuccessful in primary attendance, thereby acquiring the corresponding first face image, and further performing face recognition on the first face image through the face recognition attendance module 15, and realizing attendance checking on the attendance checking person.

The device for attendance checking by multiple persons in the embodiment of the invention utilizes the first camera to monitor the designated area, when one or more attendance persons exist in the designated area through the first camera, firstly, the first camera is used for carrying out face recognition attendance checking, then, for the attendance persons who fail to finish face recognition attendance checking preliminarily through the first camera, the corresponding second camera is opened according to the designated position of each attendance person, and each second camera is used for finishing corresponding face recognition attendance checking respectively, so that the attendance persons in the designated area can carry out face recognition attendance checking jointly, and the attendance checking efficiency based on face recognition is effectively improved.

Referring to fig. 6, the apparatus for checking attendance by multiple persons in an embodiment of the present invention further includes:

the detection module 09 is used for detecting whether an attendance checking person approaches the face recognition system;

the opening module 10 is configured to open the first camera to collect a current image of the designated area when an attendance approaches the face recognition system.

In this embodiment, specifically, the detection module 09 may detect whether an attendance is close to the face recognition system through an existing human body sensing technology, for example, an infrared sensor or a proximity sensor may be arranged on the face recognition system to detect whether an attendance is close to the face recognition system, and when the detection module 09 detects that an attendance is close to the face recognition system, the starting module 10 starts the first camera according to a signal fed back by the infrared sensor or the proximity sensor to collect a current image of the designated area, so that the current image of the designated area may be collected when the attendance is close to the face recognition system, the first camera is prevented from being continuously started, and a power saving effect is achieved; of course, the power saving effect can also be achieved by setting the timing to turn on the first camera, for example, for an enterprise heading for nine nights and six nights, the time period for turning on the first camera can be set to be 7 o 'clock to 10 o' clock in the morning and five o 'clock in the afternoon to eight o' clock in the evening.

Referring to fig. 2 and 4, the second analysis module 13 specifically includes:

the first matching unit 131 is configured to match a second face image of an attendance in the current image with a plurality of preset face images in the database, and analyze whether an attendance with an unsuccessful matching exists, where the preset face images are associated with corresponding preset user information;

the analysis unit 132 is configured to, when an attendance checking person with an unsuccessful matching exists, analyze position information of a second camera view angle range corresponding to the attendance checking person with the unsuccessful matching to form an analysis result;

the first output unit 133 is configured to output corresponding first prompt information according to preset user information when there is no attendance checking person with unsuccessful matching.

Specifically, the face recognition system can analyze a second face image existing in the current image by performing face detection on the current image, then match the detected second face image with a plurality of preset face images in the database one by one through the first matching unit 131 to obtain corresponding matching scores, and when the matching score corresponding to a certain second face image is greater than a preset threshold value, it indicates that the identity verification of the attendance corresponding to the second face image is successful, that is, there is an attendance who is successfully matched; and when the matching score corresponding to a certain second face image is smaller than a preset threshold value, the authentication of the attendance corresponding to the second face image fails, namely, the attendance with unsuccessful matching exists.

For example, referring to fig. 7 and 10, a comparison image 3 may be stored in the face recognition system in advance, where the comparison image 3 is a two-dimensional planar image corresponding to the designated area preset in the face recognition system when no attendance exists in the designated area, the comparison image 3 includes a preset area, the preset area is a position of the designated position in the comparison image 3, that is, the position of the preset area in the comparison image 3 is equivalent to the position of the designated position in the designated area, for example, designated positions a, b, and c exist in the designated area, so that three one-to-one corresponding preset areas one 31, two 32, and three 33 exist in the comparison image 2, and the preset area may be a circular area or a square area, and the specific shape may be preset by the face recognition system; when the face recognition system performs face detection (or head detection) on the current image 2, operations such as morphological processing, light compensation and the like can be performed on the current image 2 through the analysis unit 132, and a face (or head) area 21 (i.e., an area where a face exists) in the current image is extracted, wherein the face area 21 can be a circular area or a square (rectangular or square) area, and the specific shape can be preset by the face recognition system; specifically, the current image and the comparison image may be compared by the analysis unit 132, and whether the center of the face region 21 falls within the preset region is determined to determine whether the face region is located within the preset region, for example, the face region 21 is a circular region, when the center of the circular region falls within the preset region, it indicates that the face region 21 falls within the preset region, and when the center of the circular region does not fall within the preset region, it indicates that the face region 21 does not fall within the preset region; when the face area 21 falls within the preset area, it indicates that an attendance is standing at the designated position, and then according to the preset area where the face area 21 falls, the analysis unit 132 can determine the designated position where each attendance is respectively located, for example, when the current image 2 is compared with the comparison image 3 by the analysis unit 132, and two face areas 21 in the comparison image are respectively located in a first preset area 31 and a second preset area 32, it indicates that an attendance exists at the designated position a and the designated position b, and an attendance does not exist at the designated position c; therefore, the face recognition system can learn whether attendance checking persons exist in the designated area or not and positions of each attendance checking person (including attendance checking persons with unsuccessful preliminary attendance checking) in the designated area through the analysis unit 132, and then the face recognition system can conveniently start a corresponding second camera to acquire a first face image.

Before face recognition attendance, face images and user information of all attendance checking persons can be pre-recorded through a face recognition system, and a database is established, wherein the face images of the attendance checking persons correspond to the user information one by one (namely the preset face images are associated with corresponding preset user information); specifically, because the preset face images correspond to the preset user information one to one, after the attendance identity verification is successful, the first matching unit 131 can identify which attendance has finished the card punching according to the preset user information, and accordingly, the first output unit 133 can output, for example, a word or voice of "a certain person, the card punching is successful" to remind the corresponding attendance to finish the card punching.

Referring to fig. 6, the determining module 14 specifically includes:

the opening unit 141 is configured to open the second camera to acquire the first face image when the analysis result includes the position information of the second camera view angle range corresponding to the at least one attendance.

In this embodiment, specifically, the analysis result includes that each attendance has completed face recognition attendance, each attendance has not completed face recognition attendance, and some attendance has completed face recognition attendance, and specifically, when the analysis result is that each attendance has not completed face recognition attendance or some attendance has completed face recognition attendance, the corresponding analysis result further includes position information of the second camera view angle range corresponding to at least one attendance, so that the opening unit 141 subsequently opens the corresponding second camera according to the position information to acquire the first face image of the attendance with unsuccessful preliminary attendance.

Referring to fig. 2 and 3, the first analysis module 12 specifically includes:

a head detection unit 121, configured to perform head detection on the current image to obtain head information in the current image, where the head information includes the number of heads;

a first judging unit 122, configured to judge whether the number of heads is greater than or equal to a preset value;

and the judging unit 123 is configured to judge that one or more attendance checking persons exist in the designated area when the number of heads is greater than or equal to a preset value.

Specifically, the human head detecting unit 121 receives a current image from the first camera and performs human head detection on the current image by using an existing human head detecting algorithm, so as to obtain human head information in the current image, where the human head information includes the number of human heads, so as to perform related operations in the following.

The first determining unit 122 compares the number of the human heads in the obtained human head information with a preset value, and determines whether the number of the human heads in the current image is greater than or equal to the preset value, wherein the preset value is greater than or equal to 1.

If the number of people in the current image is greater than or equal to the preset value, it indicates that one or more attendance checking persons exist in the designated area, and therefore the determination unit 123 may determine that one or more attendance checking persons exist in the designated area accordingly.

Referring to fig. 2 and 5, the first face images are multiple, and the face recognition attendance module 15 specifically includes:

a configuration unit 151 configured to configure a designated identifier for each of the first face images, respectively;

a second matching unit 152, configured to perform similarity matching between each first face image configured with a designated identifier and a plurality of preset face images in a database, to obtain a preset face image with the maximum similarity to the first face image, where the preset face image is associated with corresponding preset user information;

a second judging unit 153, configured to judge whether a similarity between the first face image and the preset face image is greater than a preset threshold;

and a second output unit 154, configured to output corresponding first prompt information according to preset user information when the similarity between the first face image and the preset face image is greater than a preset threshold.

And the third output unit 155 is configured to output corresponding second prompt information according to the specified identifier on the first face image when the similarity between the first face image and the preset face image is not greater than the preset threshold.

Specifically, for example, after the configuration unit 151 acquires a first face image captured by the second camera a, a specific identifier, for example, the number "1", may be configured for the first face image, and similarly, a specific identifier, for example, the number "2", may be configured for the first face image captured by the second camera B; a designated identifier, for example, the number "3", may be configured for the first face image captured by the second camera C, so that the first face image captured by each second camera may be distinguished by the designated identifier, that is, the identifier may be used to determine from which designated position the first face image is from the attendance checker, so as to perform subsequent operations.

Before face recognition attendance, face images and user information of all attendance checking persons are recorded in advance through a face recognition system, and a database is established, wherein the face images of the attendance checking persons correspond to the user information one by one (namely the preset face images are associated with corresponding preset user information); during verification, the second matching unit 152 performs similarity matching between the acquired first face images and the preset face images in the database one by one, so as to search a corresponding matching image with the highest similarity from the database, for example, the second matching unit 152 first obtains a first face image "1" from the second camera a, then the second matching unit 152 performs similarity matching between the first face image "1" and the preset face images in the database one by one, so as to obtain a plurality of similarities, and then screens out a preset face image with the highest similarity, the preset face image is a face image preliminarily matched with the first face image "1" in the database, and when the first face image "2" from the second camera B or the first face image "3" from the second camera C is obtained, the process of similarity matching by the second matching unit 152 is similar to the above, this will not be described in detail.

When the matched preset face image is obtained, the second judging unit 153 compares the similarity between the first face image and the preset face image with a preset threshold, judges whether the similarity is greater than the preset threshold, if so, indicates that the first face image obtained by the face recognition system and the preset face image in the database are both from the same person, and the attendance identity verification is successful.

The first prompt information may be presented in a text form, or in a voice form, or in a combination form of text and voice, specifically, since the face image preset in the database by the attendance is in one-to-one correspondence with the user information, after the attendance is successfully authenticated, the second matching unit 152 may identify which attendance has finished the card punching according to the user information, and at this time, the second output unit 154 may output, for example, a word "a certain person, the card punching has been successfully" or a voice to remind the corresponding attendance that the card punching has been finished.

Specifically, if the similarity between the first face image and the preset face image is not greater than the preset threshold, it indicates that the first face image acquired by the second matching unit 152 and the preset face image in the database are not from the same person, and the attendance checking authentication fails, at this time, since the first face images acquired by the second matching unit 152 are all configured with the specified identifier, the second matching unit 152 may identify which attendance checking person at the specified position fails to be successfully checked according to the specified identifier, for example, when the first face image "1" acquired by the first camera a fails to be successfully matched, the second matching unit 152 may determine that the first face image failed to be successfully matched is from the attendance checking person at the position a according to the specified identifier "1", and at this time, the third output unit 155 may output, for example, "position a fails to be checked, please re-try the words or voices to remind the corresponding attendance to punch the card again.

Referring to fig. 2 and 5, the face recognition attendance module 15 specifically further includes:

an obtaining unit 156, configured to obtain a current attendance time;

a third judging unit 157, configured to judge whether the current attendance time exceeds a preset time interval;

the first generating unit 158 is configured to generate corresponding first attendance information according to the current attendance time and preset user information when the current attendance time does not exceed the preset time interval;

and a second generating unit 159, configured to generate corresponding second attendance information according to the current attendance time and preset user information when the current attendance time exceeds the preset time interval.

The preset user information includes personal information such as name, department, and position of the attendance, and after the attendance completes the identity verification, the obtaining unit 156 obtains the current attendance time of the attendance so as to perform the related operations subsequently.

The preset time interval may be determined according to actual conditions, for example, for a business facing nine nights and six nights, the preset time interval may be set to seven to nine am and six to eight pm.

If the third determining unit 157 determines that the current attendance time does not exceed the preset time interval, it indicates that the attendance person completed attendance within the specified time, at this time, the first attendance information generated by the first generating unit 158 is normal, that is, the first attendance information is displayed with an evaluation result in addition to the personal information and attendance time of the attendance person, and the evaluation result can be displayed through a word of "normal attendance", so that the manager can quickly know the attendance result of the attendance person, and the manager can conveniently count the attendance condition of the attendance person.

If the third determining unit 157 determines that the current attendance time exceeds the preset time interval, it indicates that the attendance person completed attendance within the unspecified time, and at this time, the second attendance information generated by the second generating unit 159 is abnormal, that is, the second attendance information displays an evaluation result in addition to the personal information and the attendance time of the attendance person, and the evaluation result can be displayed through a word of "attendance abnormal", so that the manager can quickly know the attendance result of the attendance person, and the manager can conveniently count the attendance condition of the attendance person.

Referring to fig. 6, the apparatus for checking attendance by multiple persons in the embodiment of the present invention further includes:

and the replacing module 16 is used for replacing the preset face image matched with the first face image in the database with the first face image.

In this embodiment, when a certain first face image acquired by the face recognition attendance module 15 is successfully matched with a certain preset face image in the preset database, the preset face image is replaced by the first face image through the replacement module 16 to update the preset face image corresponding to the attendance in the database, and since the longer the time interval is, the larger the difference between the acquired first face image and the preset face image in the database may be, the case that the identity of the attendance is true may be caused to cause normal attendance, the replacement module 16 dynamically updates the preset face image corresponding to the attendance in the database, which is beneficial to reducing the risk of face image verification failure due to the overlong time interval, and thus improving the stability of face recognition attendance.

the counting module 17 is used for counting the frequency of face recognition attendance failure in the attendance time period and judging whether the frequency is greater than the preset frequency or not;

the storage module 18 is configured to store the video information acquired by the first camera in the attendance time period when the number of times of face recognition attendance failure in the attendance time period is greater than a preset number of times.

In this embodiment, specifically, the first camera may adopt a monitoring camera having a real-time video recording function and a shooting function on the market, and in the specific implementation, the video information of the designated area may be collected by setting the real-time video recording function of the first camera to be started at regular time, for example, for an enterprise facing nine nights and six nights, the time period for the first camera to start the real-time video recording function may be set to be 7 am to 10 am, and five pm to eight pm; meanwhile, whether an attendance person approaches the face recognition system or not can be detected by arranging an infrared sensor or a proximity sensor on the face recognition system, and when the attendance person is detected to approach the face recognition system, the face recognition system starts a shooting function of a first camera according to a signal fed back by the infrared sensor or the proximity sensor to acquire a current image of a designated area; specifically, after the face recognition attendance module 15 performs face recognition attendance respectively according to the first face images acquired by the second cameras, the number of face recognition attendance failures in the attendance time period can be counted by the counting module 17, for example, if the morning attendance time period is from 7 to 10 am, the number of face recognition attendance failures in the time period can be counted before 10 am, wherein the reason of the face recognition attendance failures may be that the attendance is cheated (for example, cheated by a photo), or a preset threshold (for verification of the attendance identity, which is consistent with the above) is set too high; when the counting module 17 judges that the number of times of face recognition attendance failure is greater than the preset number of times, the video information acquired by the first camera in the morning time period is stored through the storage module 18, wherein the preset number of times can be determined according to the actual conditions of enterprises, so that whether cheating occurs to an attendance person can be known by checking the video information, if the cheating phenomenon is not found through the video information, the reason that the preset threshold value is set too high is probably shown, and at the moment, a manager can appropriately adjust the preset threshold value built in the face recognition system; when the number of times of face recognition attendance failure is judged to be not more than the preset number of times, the storage module 18 does not store the video information acquired by the first camera in the morning time period so as to save the storage space of the face recognition system; therefore, in the implementation, the purpose of face recognition attendance can be achieved, the attendance condition in the designated area can be monitored in the attendance time period through the first camera, so that the occurrence of the phenomenon that an attendance person replaces attendance can be reduced without increasing a living body detection technology, meanwhile, the preset threshold value can be practically adjusted in a matching manner, and the flexibility and the practicability of the face recognition system are improved.

The embodiment of the present invention further provides a face recognition system, which includes a first camera, a second camera, a memory, a processor, and an application program, where the first camera and the second camera are respectively connected to the processor, the application program is stored in the memory and configured to be executed by the processor, and the application program is configured to be used for executing the method for checking attendance by multiple people in any of the above embodiments.

It will be appreciated by those skilled in the art that the face recognition system according to the embodiments of the present invention is an apparatus as described above for performing one or more of the methods described in the present application. These devices may be specially designed and manufactured for the required purposes, or they may comprise known devices in general-purpose computers. These devices have stored therein computer programs or applications that are selectively activated or reconfigured. Such a computer program may be stored in a device (e.g., computer) readable medium, including, but not limited to, any type of disk including floppy disks, hard disks, optical disks, CD-ROMs, and magnetic-optical disks, ROMs (Read-Only memories), RAMs (Random Access memories), EPROMs (Erasable Programmable Read-Only memories), EEPROMs (Electrically Erasable Programmable Read-Only memories), flash memories, magnetic cards, or optical cards, or any type of media suitable for storing electronic instructions, and each coupled to a bus. That is, a readable medium includes any medium that stores or transmits information in a form readable by a device (e.g., a computer).

The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.

Claims

1. The utility model provides a method for attendance is carried out jointly to many people, its characterized in that is applied to face identification system, face identification system includes first camera and a plurality of second camera, wherein, the visual angle scope of first camera covers the appointed area, and each the visual angle scope of second camera covers the assigned position respectively, the assigned position cover in the appointed area, each the second camera can set up according to certain angle or/and each second camera between the certain distance of interval set up, make each the second camera cover respectively the assigned position, the method includes following step:

acquiring a current image of a designated area acquired by the first camera;

if one or more attendance checking persons exist in the specified area, analyzing the attendance checking persons to form an analysis result; the analysis result comprises that each attendance checker completes face recognition attendance, each attendance checker does not complete face recognition attendance and part of the attendance checkers complete face recognition attendance;

determining whether to start the second camera according to the analysis result so as to acquire a first face image;

performing face recognition attendance according to the first face image;

the step of determining whether to turn on the second camera according to the analysis result to acquire a first face image includes:

if the analysis result indicates that each attendance checker does not finish face recognition attendance checking or only part of attendance checks finish face recognition attendance checking, the second cameras covering the designated positions are started according to the designated positions of the attendance checkers, and the second cameras finish corresponding face recognition attendance checking respectively so as to acquire first face images of attendance checkers with unsuccessful preliminary attendance checking, so that a plurality of attendance checkers in the designated area carry out face recognition attendance checking together.

2. The method for attendance checking by multiple persons in combination according to claim 1, wherein the step of analyzing the attendance checking person to form an analysis result comprises:

matching a second face image of the attendance in the current image with a plurality of preset face images in a database, and analyzing whether an attendance with unsuccessful matching exists or not;

3. The method for attendance checking by multiple persons in combination according to claim 1, wherein the step of determining whether to turn on the second camera for collecting the first face image according to the analysis result comprises:

4. The method for attendance checking by multiple persons in combination according to claim 1, wherein the step of analyzing whether one or more attendance checking persons exist in the designated area or not according to the current image comprises the following steps:

and if the number of the heads is larger than or equal to a preset value, judging that one or more attendance checking persons exist in the designated area.

5. The method for attendance checking by multiple persons according to any one of claims 1 to 4, wherein the first face image is multiple, and the step of performing face recognition attendance checking according to the first face image comprises:

respectively configuring a designated identifier for each first face image;

judging whether the similarity between the first face image and the preset face image is greater than a preset threshold value or not;

and if so, outputting corresponding first prompt information according to the preset user information.

6. The method for attendance checking by multiple persons in combination according to claim 5, wherein after the step of judging whether the similarity between the first face image and the preset face image is greater than a preset threshold value, the method further comprises:

and if the similarity between the first face image and the preset face image is not greater than a preset threshold value, outputting corresponding second prompt information according to the designated identifier on the first face image.

7. The method for attendance checking by multiple persons in combination according to claim 5, wherein after the step of outputting the corresponding first prompt information according to the preset user information, the method further comprises:

8. The method for attendance checking by multiple persons in combination according to any one of claims 1 to 4, wherein after the step of performing face recognition attendance checking according to the first face image, the method further comprises:

counting the frequency of face recognition attendance failure in an attendance time period, and judging whether the frequency is greater than a preset frequency or not;

9. The utility model provides a device that many people carry out attendance jointly which characterized in that is applied to face recognition system, face recognition system includes first camera and a plurality of second camera, wherein, the visual angle scope of first camera covers the specified area, each the visual angle scope of second camera covers the assigned position respectively, the assigned position cover in the specified area, each the second camera can set up according to certain angle or/and each second camera between certain distance of interval set up, make each the second camera covers respectively the assigned position, the device includes:

the second analysis module is used for analyzing one or more attendance checking persons to form an analysis result when the attendance checking persons exist in the specified area; the analysis result comprises that each attendance checker completes face recognition attendance, each attendance checker does not complete face recognition attendance and part of the attendance checkers complete face recognition attendance;

the determining module is used for opening the second cameras covering the designated positions according to the designated positions of the attendance checking persons if the analysis result indicates that the face recognition attendance checking of all the attendance checking persons is not completed or only part of the attendance checking persons are completed, and completing the corresponding face recognition attendance checking by the second cameras respectively so as to acquire first face images of the attendance checking persons with unsuccessful primary attendance checking and further enable a plurality of attendance checking persons in the designated area to perform the face recognition attendance checking together;

10. A face recognition system comprising a first camera, a second camera, and a memory, a processor, and an application, the first camera and the second camera being respectively connected to the processor, the application being stored in the memory and configured to be executed by the processor, the application being configured to perform the method of attendance checking by multiple persons according to any one of claims 1 to 8.