One-key roll call system based on face recognition

Abstract

The invention relates to the technical field of biological recognition, in particular to a one-key roll call system based on face recognition. According to the technical scheme, the one-key roll call system based on face recognition comprises a face registration module, a face recognition module and a one-key roll call module; compared with a traditional one-key roll call system, conference sign-in generally adopts a manual mode, workers need to check identity information of attendance personnel and sign in one by one, time and labor are wasted, the labor cost is high, card swiping or code scanning sign-in modes are adopted in some occasions, however, each participant needs to perform active operation, and the work efficiency is high. Due to the fact that manual operation is prone to omission and missed signing is caused, the one-key roll call system utilizes the face recognition technology, the system can quickly recognize identities of participants, sign-in of the whole meeting place can be completed at a time, one-by-one operation of the participants is not needed, meanwhile, all the participants in the meeting place are automatically collected and recognized through the camera, omission is not prone to occurring, and the work efficiency is improved. And the missed signing risk in the traditional signing mode is reduced.

Description

Technical Field

[0001] The present invention relates to the field of biometric identification technology, and in particular to a one-key roll call system based on face recognition. Background Art

[0002] The one-click roll call system is a fast roll call tool that uses modern technology. Through the software platform, the system allows teachers to quickly and accurately complete student roll calls in the classroom or at activities, and record relevant attendance information.

[0003] The existing one-click roll call system usually uses manual methods to sign in meetings, requiring staff to verify the identity information of the attendees and sign in one by one, which is time-consuming and labor-intensive, and has high labor costs. At the same time, some occasions also use card swiping or code scanning to sign in, but this requires each participant to take the initiative to operate, which is inefficient and prone to omissions due to manual operation.

[0004] The existing one-click roll call system usually uses manual methods to sign in for meetings. Staff members need to verify the identity information of the attendees and sign in one by one, which is time-consuming and labor-intensive, and has high labor costs. The one-click roll call system uses facial recognition technology. The system can quickly identify the identities of the participants and complete the sign-in of the entire venue at one time. There is no need for staff to operate one by one. At the same time, the camera automatically collects and identifies all participants in the venue, making it difficult to miss them, reducing the risk of missed signatures under traditional sign-in methods. Summary of the invention

[0005] In order to overcome the existing one-click roll call system, meeting sign-in is usually done manually, requiring staff to verify the identity information of the attendees and sign in one by one, which is time-consuming and labor-intensive, and has high labor costs. At the same time, some occasions also use card swiping or code scanning to sign in, but this requires each participant to take the initiative to operate, which is inefficient and prone to omissions due to manual operation, resulting in missed signatures.

[0006] The technical solution of the present invention is: a one-key roll call system based on face recognition, comprising:

[0007] Face registration module: used to collect face data of participants and establish a face database;

[0008] Face recognition module: used to analyze and identify face images through face recognition algorithms;

[0009] One-click roll call module: used for users to perform one-click roll call operation;

[0010] Attendance record storage and query module: used to store and manage the attendance records of each meeting;

[0011] Sign-in record manual review module: used for subsequent review and modification of meeting sign-in lists.

[0012] Preferably, the face registration module is used to collect the face data of the participants and establish a face database, obtain the face images of the participants through a mobile phone APP or other image acquisition devices, and store these face images in association with the corresponding personal identity information, such as name, work number, etc., to ensure the uniqueness of personal information, build a face database, and lay the foundation for subsequent face recognition and comparison.

[0013] Preferably, the face registration module comprises the following steps when working:

[0014] S11: Acquire facial images of participants through a camera or other image acquisition device;

[0015] S12: preprocessing the acquired face image, denoising the image, and enhancing the contrast to improve the subsequent face detection and recognition accuracy;

[0016] S13: Face detection is based on the YOLOV8 algorithm and modified to use the ArcFaceLoss loss function training to detect and locate the face area and the five facial key points in the preprocessed image;

[0017] S14: Using the face feature extraction algorithm Resnet50 to extract the face feature value of the detected face area into a feature vector with a dimension of 512;

[0018] S15: input the identity information of the participants, such as name, work number, etc., and associate the extracted facial feature values ​​with the corresponding identity information and store them in the face database;

[0019] S16: Repeat steps S11-S15 until the face registration of all participants is completed.

[0020] Preferably, the face recognition module is used to collect images of the venue in real time through cameras in the venue at the beginning of the meeting, and use a face recognition algorithm to detect the face of each participant from the collected image, and then compare the detected face image with the face data in the previously established face library to determine the identity information corresponding to each recognized face.

[0021] Preferably, when the face recognition module is working, the following steps are included:

[0022] S21: Collecting images of the venue in real time through the cameras in the venue, and passing the images to the face recognition algorithm module;

[0023] S22: Preprocess the collected image, remove noise and enhance contrast of the image, and the preprocessing method is consistent with the face registration module;

[0024] S23: applying a face detection algorithm to detect all face regions and face key point regions in the image, and mapping the face to a standard space through affine transformation, so as to achieve higher recognition for different angles and expression changes;

[0025] S24: extracting facial feature values ​​for each detected facial region using a feature extraction algorithm;

[0026] S25: Compare the extracted facial feature values ​​with the feature values ​​in a pre-established facial database to calculate a similarity score;

[0027] S26: According to the set similarity threshold, the face with a score higher than the threshold is matched with the identity information in the database to determine the identity corresponding to the face;

[0028] S27: Repeat steps S24-S26 until all faces in the venue are identified;

[0029] S28: Summarize the identity information of all identified personnel and output it as the attendance list for this meeting.

[0030] Preferably, the one-click roll call module is used to provide an operating interface for one-click roll call. When the venue is equipped with a display or intelligent conference system, the conference administrator only needs to click the "one-click roll call" button on the interface, and the system can automatically collect pictures of the meeting site and perform facial comparison on all people on site. According to the identity information obtained by the previous registration module, a complete sign-in list for this meeting is automatically generated, without the need for manual roll call one by one, thus realizing true one-click sign-in.

[0031] Preferably, the one-key roll call module comprises the following steps when working:

[0032] S31: Detecting that the "One-key roll call" button is clicked;

[0033] S32: calling the face recognition module to perform face recognition of the participants at the meeting and obtain the attendance list of the meeting;

[0034] S33: Display the check-in progress and results in real time on the interface;

[0035] S34: After the sign-in is completed, the final sign-in list is displayed;

[0036] S35: Upload the current sign-in information to the sign-in record storage module.

[0037] Preferably, the sign-in record storage and query module is used to store and manage the sign-in records of each meeting. The system will store the sign-in list generated by the one-click roll call operation, as well as relevant data such as sign-in time and meeting information into the database. Administrators can easily query historical sign-in records for subsequent review or statistics.

[0038] Preferably, the sign-in record storage and query module includes the following steps when working:

[0039] S41: receiving the sign-in information transmitted from the one-key roll call module, including the sign-in list, meeting information, topic, time, location, etc.;

[0040] S42: Establish a database record, which includes two parts: (1) meeting information, such as meeting topic, time, location, etc.; (2) attendance data of participants, including name, position, attendance time, etc.;

[0041] S43: storing the newly created record in the database;

[0042] S44: Provides a database query interface, where administrators can query historical sign-in records as needed, and supports retrieval and screening by meeting topic, time period, attendees, etc.

[0043] S45: Display the query results in a table or other form on the interface.

[0044] As a preference, the manual review module for sign-in records is used for subsequent review and modification of the meeting sign-in list. If there are people who are misidentified by the algorithm, whose faces are not registered in the system, or who fail to arrive at the venue at the sign-in time, the meeting administrator can correct the sign-in list in the corresponding module of the meeting management platform, thus achieving humanized sign-in on an intelligent basis.

[0045] Preferably, the one-click roll call system has a fast sign-in speed and improves the sign-in efficiency. By utilizing face recognition technology, the system can quickly identify the identities of participants and complete the sign-in of the entire venue at one time, without the need for personnel to operate one by one, which greatly improves the sign-in efficiency. At the same time, it has high accuracy and reduces the risk of missed sign-ins and proxy sign-ins. The system automatically collects and identifies all participants in the venue through cameras, which is not easy to miss, reducing the risk of missed sign-ins under traditional sign-in methods. Each face has corresponding feature information, reducing the problem of someone signing in on behalf of others in traditional sign-ins.

[0046] Beneficial effects of the present invention:

[0047] 1. Compared with the traditional one-click roll call system, conference sign-in is usually done manually, requiring staff to verify the identity information of the attendees and sign in one by one, which is time-consuming and labor-intensive, and has high labor costs. At the same time, some occasions also use card swiping or code scanning to sign in, but this requires each participant to take the initiative to operate, which is inefficient, and it is easy to miss due to manual operation, resulting in missed signatures. The one-click roll call system uses face recognition technology. The system can quickly identify the identity of the participants and complete the sign-in of the entire venue at one time without the need for personnel to operate one by one. At the same time, the camera automatically collects and identifies all participants in the venue, which is not easy to miss, reducing the risk of missed signatures under traditional sign-in methods;

[0048] 2. The one-click roll call system has a fast sign-in speed and improves the sign-in efficiency. By using face recognition technology, the system can quickly identify the identities of participants and complete the sign-in of the entire venue at one time. There is no need for personnel to operate one by one, which greatly improves the sign-in efficiency. At the same time, it has high accuracy and reduces the risk of missed sign-ins and proxy sign-ins. The system automatically collects and identifies all participants in the venue through cameras, which is not easy to miss, reducing the risk of missed sign-ins under traditional sign-in methods. Each face has corresponding feature information, which reduces the problem of someone signing in on behalf of others in traditional sign-ins. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 What is shown is a schematic diagram of a one-key roll call system framework based on face recognition of the present invention;

[0050] Figure 2 What is shown is a schematic diagram of the working process of a face registration module of a one-key roll-call system based on face recognition of the present invention;

[0051] Figure 3 What is shown is a schematic diagram of the working process of a face recognition module of a one-key roll-call system based on face recognition of the present invention;

[0052] Figure 4 What is shown is an interface effect diagram of a one-key roll-call module of a one-key roll-call system based on face recognition of the present invention;

[0053] Figure 5 What is shown is an interface effect diagram of a sign-in record storage and review module of a one-key roll call system based on face recognition according to the present invention. DETAILED DESCRIPTION

[0054] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0055] See also Figure 1-Figure 4 The present invention provides an embodiment: a one-key roll call system based on face recognition, comprising:

[0056] Face registration module: used to collect face data of participants and establish a face database;

[0057] Face recognition module: used to analyze and identify face images through face recognition algorithms;

[0058] One-click roll call module: used for users to perform one-click roll call operation;

[0059] Attendance record storage and query module: used to store and manage the attendance records of each meeting;

[0060] Sign-in record manual review module: used for subsequent review and modification of meeting sign-in lists.

[0061] Preferably, the face registration module is used to collect the face data of the participants and establish a face database, obtain the face images of the participants through a mobile phone APP or other image acquisition devices, and store these face images in association with the corresponding personal identity information, such as name, work number, etc., to ensure the uniqueness of personal information, build a face database, and lay the foundation for subsequent face recognition and comparison.

[0062] Preferably, the face registration module comprises the following steps when working:

[0063] S11: Acquire facial images of participants through a camera or other image acquisition device;

[0064] S12: preprocessing the acquired face image, denoising the image, and enhancing the contrast to improve the subsequent face detection and recognition accuracy;

[0065] S13: Face detection is based on the YOLOV8 algorithm and modified to use the ArcFaceLoss loss function training to detect and locate the face area and the five facial key points in the preprocessed image;

[0066] S14: Using the face feature extraction algorithm Resnet50 to extract the face feature value of the detected face area into a feature vector with a dimension of 512;

[0067] S15: input the identity information of the participants, such as name, work number, etc., and associate the extracted facial feature values ​​with the corresponding identity information and store them in the face database;

[0068] S16: Repeat steps S11-S15 until the face registration of all participants is completed.

[0069] Preferably, the face recognition module is used to collect images of the venue in real time through cameras in the venue at the beginning of the meeting, and use a face recognition algorithm to detect the face of each participant from the collected image, and then compare the detected face image with the face data in the previously established face library to determine the identity information corresponding to each recognized face.

[0070] Preferably, when the face recognition module is working, the following steps are included:

[0071] S21: Collecting images of the venue in real time through the cameras in the venue, and passing the images to the face recognition algorithm module;

[0072] S22: Preprocess the collected image, remove noise and enhance contrast of the image, and the preprocessing method is consistent with the face registration module;

[0073] S23: applying a face detection algorithm to detect all face regions and face key point regions in the image, and mapping the face to a standard space through affine transformation, so as to achieve higher recognition for different angles and expression changes;

[0074] S24: extracting facial feature values ​​for each detected facial region using a feature extraction algorithm;

[0075] S25: Compare the extracted facial feature values ​​with the feature values ​​in a pre-established facial database to calculate a similarity score;

[0076] S26: According to the set similarity threshold, the face with a score higher than the threshold is matched with the identity information in the database to determine the identity corresponding to the face;

[0077] S27: Repeat steps S24-S26 until all faces in the venue are identified;

[0078] S28: Summarize the identity information of all identified personnel and output it as the attendance list for this meeting.

[0079] Preferably, the one-click roll call module is used to provide an operating interface for one-click roll call. When the venue is equipped with a display or intelligent conference system, the conference administrator only needs to click the "one-click roll call" button on the interface, and the system can automatically collect pictures of the meeting site and perform facial comparison on all people on site. According to the identity information obtained by the previous registration module, a complete sign-in list for this meeting is automatically generated, without the need for manual roll call one by one, thus realizing true one-click sign-in.

[0080] Preferably, the one-key roll call module comprises the following steps when working:

[0081] S31: Detecting that the "One-key roll call" button is clicked;

[0082] S32: calling the face recognition module to perform face recognition of the participants at the meeting and obtain the attendance list of the meeting;

[0083] S33: Display the check-in progress and results in real time on the interface;

[0084] S34: After the sign-in is completed, the final sign-in list is displayed;

[0085] S35: Upload the current sign-in information to the sign-in record storage module.

[0086] Preferably, the sign-in record storage and query module is used to store and manage the sign-in records of each meeting. The system will store the sign-in list generated by the one-click roll call operation, as well as relevant data such as sign-in time and meeting information into the database. Administrators can easily query historical sign-in records for subsequent review or statistics.

[0087] Preferably, the sign-in record storage and query module includes the following steps when working:

[0088] S41: receiving the sign-in information transmitted from the one-key roll call module, including the sign-in list, meeting information, topic, time, location, etc.;

[0089] S42: Establish a database record, which includes two parts: (1) meeting information, such as meeting topic, time, location, etc.; (2) attendance data of participants, including name, position, attendance time, etc.;

[0090] S43: storing the newly created record in the database;

[0091] S44: Provides a database query interface, where administrators can query historical sign-in records as needed, and supports retrieval and screening by meeting topic, time period, attendees, etc.

[0092] S45: Display the query results in a table or other form on the interface.

[0093] As a preference, the manual review module for sign-in records is used for subsequent review and modification of the meeting sign-in list. If there are people who are misidentified by the algorithm, whose faces are not registered in the system, or who fail to arrive at the venue at the sign-in time, the meeting administrator can correct the sign-in list in the corresponding module of the meeting management platform, thus achieving humanized sign-in on an intelligent basis.

[0094] Preferably, the one-click roll call system has a fast sign-in speed and improves the sign-in efficiency. By utilizing face recognition technology, the system can quickly identify the identities of participants and complete the sign-in of the entire venue at one time, without the need for personnel to operate one by one, which greatly improves the sign-in efficiency. At the same time, it has high accuracy and reduces the risk of missed sign-ins and proxy sign-ins. The system automatically collects and identifies all participants in the venue through cameras, which is not easy to miss, reducing the risk of missed sign-ins under traditional sign-in methods. Each face has corresponding feature information, reducing the problem of someone signing in on behalf of others in traditional sign-ins.

[0095] Example 1

[0096] An internal meeting sign-in application for a company

[0097] The system was applied to the internal meeting management of an enterprise. The facial registration of all employees was completed in advance through the facial acquisition equipment, and an employee facial database was established. When the meeting starts, just click the one-click roll call button, and the system will collect the on-site image through the camera in the conference room, and compare it with the facial database, automatically identify and confirm the identity of the participants, and finally generate a complete sign-in list. This sign-in system replaces the traditional manual sign-in or card sign-in, greatly improving the sign-in efficiency and effectively avoiding the risk of missed sign-ins. At the same time, the system can automatically store the sign-in records of each meeting, and managers can check them at any time according to needs, greatly reducing the management burden.

[0098] Example 2

[0099] Class attendance application in a certain university

[0100] The system is applied to the classroom teaching management of a certain university for students' classroom attendance. At the beginning of the semester, all students' facial registration is completed through camera acquisition. At the beginning of each class, the teacher only needs to click the sign-in button, and the system can automatically identify and confirm the identity of the students in class based on the on-site images captured by the camera in the classroom, and generate a sign-in list for the class. This one-click roll call method makes classroom attendance more efficient than ever before. There is no need for students to sign in one by one, which saves time and prevents problems such as teacher omissions and student absenteeism. At the same time, the sign-in records stored in the system are also convenient for teachers to review and count, improving the level of teaching management.

[0101] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge scope of those skilled in the art without departing from the spirit of the present invention.

Claims

1. A one-key roll call system based on face recognition; characterized by: Included are: Face registration module: used to collect face data of participants and establish a face database; Face recognition module: used to analyze and identify face images through face recognition algorithms; One-click roll call module: used for users to perform one-click roll call operation; Attendance record storage and query module: used to store and manage the attendance records of each meeting; Sign-in record manual review module: used for subsequent review and modification of meeting sign-in lists.

2. The one-key roll call system based on face recognition according to claim 1, characterized in that: The face registration module is used to collect the face data of the participants and establish a face database. It obtains the face images of the participants through mobile phone APP or other image acquisition devices, and stores these face images in association with the corresponding personal identity information, such as name, work number, etc., to ensure the uniqueness of personal information, build a face database, and lay the foundation for subsequent face recognition and comparison.

3. The one-key roll call system based on face recognition according to claim 2 is characterized in that: When the face registration module is working, it includes the following steps: S11: Acquire facial images of participants through a camera or other image acquisition device; S12: preprocessing the acquired face image, denoising the image, and enhancing the contrast to improve the subsequent face detection and recognition accuracy; S13: Face detection is based on the YOLOV8 algorithm and modified to use the ArcFaceLoss loss function training to detect and locate the face area and the five facial key points in the preprocessed image; S14: Using the face feature extraction algorithm Resnet50 to extract the face feature value of the detected face area into a feature vector with a dimension of 512; S15: input the identity information of the participants, such as name, work number, etc., and associate the extracted facial feature values ​​with the corresponding identity information and store them in the face database; S16: Repeat steps S11-S15 until the face registration of all participants is completed.

4. The one-key roll call system based on face recognition according to claim 1, characterized in that: The face recognition module is used to collect images of the venue in real time through cameras in the venue at the beginning of the meeting, and use the face recognition algorithm to detect the face of each participant from the collected images, and then compare the detected face images with the face data in the previously established face database to determine the identity information corresponding to each recognized face.

5. The one-key roll call system based on face recognition according to claim 4 is characterized in that: When the face recognition module is working, it includes the following steps: S21: Collecting images of the venue in real time through the cameras in the venue, and passing the images to the face recognition algorithm module; S22: Preprocess the collected image, remove noise and enhance contrast of the image, and the preprocessing method is consistent with the face registration module; S23: applying a face detection algorithm to detect all face regions and face key point regions in the image, and mapping the face to a standard space through affine transformation, so as to achieve higher recognition for different angles and expression changes; S24: extracting facial feature values ​​for each detected facial region using a feature extraction algorithm; S25: Compare the extracted facial feature values ​​with the feature values ​​in a pre-established facial database to calculate a similarity score; S26: According to the set similarity threshold, the face with a score higher than the threshold is matched with the identity information in the database to determine the identity corresponding to the face; S27: Repeat steps S24-S26 until all faces in the venue are identified; S28: Summarize the identity information of all identified personnel and output it as the attendance list for this meeting.

6. The one-key roll call system based on face recognition according to claim 1, characterized in that: The one-click roll call module is used to provide an operation interface for one-click roll call. When the venue is equipped with a display or intelligent conference system, the conference administrator only needs to click the "one-click roll call" button on this interface, and the system can automatically collect pictures of the meeting site and perform facial comparison on all people on site. According to the identity information obtained by the previous registration module, a complete sign-in list for this meeting is automatically generated, without the need for manual roll call one by one, thus realizing true one-click sign-in.

7. The one-key roll call system based on face recognition according to claim 6, characterized in that: The one-key roll call module includes the following steps when working: S31: Detecting that the "One-key roll call" button is clicked; S32: calling the face recognition module to perform face recognition of the participants at the meeting and obtain the attendance list of the meeting; S33: Display the check-in progress and results in real time on the interface; S34: After the sign-in is completed, the final sign-in list is displayed; S35: Upload the current sign-in information to the sign-in record storage module.

8. The one-key roll call system based on face recognition according to claim 1, characterized in that: The sign-in record storage and query module is used to store and manage the sign-in records of each meeting. The system will store the sign-in list generated by the one-click roll call operation, as well as relevant data such as sign-in time and meeting information in the database. Administrators can easily query historical sign-in records for subsequent review or statistics.

9. The one-key roll call system based on face recognition according to claim 8, characterized in that: The sign-in record storage and query module includes the following steps when working: S41: receiving the sign-in information transmitted from the one-key roll call module, including the sign-in list, meeting information, topic, time, location, etc.; S42: Establish a database record, which includes two parts: (1) meeting information, such as meeting topic, time, location, etc.; (2) attendance data of participants, including name, position, attendance time, etc.; S43: storing the newly created record in the database; S44: Provides a database query interface, where administrators can query historical sign-in records as needed, and supports retrieval and screening by meeting topic, time period, attendees, etc. S45: Display the query results in a table or other form on the interface.

10. The one-key roll call system based on face recognition according to claim 9, characterized in that: The manual review module of sign-in records is used for subsequent review and modification of the meeting sign-in list. If there are people who are misidentified by the algorithm, whose faces are not registered in the system, or who fail to arrive at the venue at the sign-in time, the conference administrator can correct the sign-in list in the corresponding module of the conference management platform, so as to achieve humanized sign-in on the basis of intelligence.

Images