Face recognition access control system

Abstract

The invention relates to the technical field of intelligent access control, and discloses a face recognition access control system, which comprises a personnel information collection module, a data security and privacy protection module, a face and real scene detection module, a feature extraction module, a face comparison module and a system management module. The face recognition access control system plays a deterrent role in potential lawbreakers, reduces the probability of occurrence of illegal intrusion events, can avoid crowding and pushing caused by simultaneous access control of multiple persons in a crowded occasion, effectively prevents stampede accidents, guarantees the life safety of the persons, and meanwhile, when a safety problem occurs, the face recognition access control system can effectively prevent stampede accidents from happening. The system records multi-person passing information and early warning, the event process can be traced conveniently, a powerful basis is provided for follow-up investigation, finally, for normally-passing persons, the early warning function of the system can enable the persons to feel strict safety management of a place, and the sense of security and comfort of the persons in the place are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent access control, and in particular to a face recognition access control system. Background Art

[0002] Facial recognition technology is a technology that uses facial images as features and utilizes image processing and pattern recognition technologies to identify and authenticate faces. Access control gates recognize faces based on facial recognition technology. It uses a camera to capture facial images, uses facial recognition algorithms to extract features, and compares them with facial data stored in the database to determine whether someone has permission to enter or pass through the gate.

[0003] Facial recognition access control has been widely used in our daily life, such as gates at office buildings, community gates, and apartment building gates. The purpose of setting up facial recognition access control is to identify people entering and leaving the community, prevent strangers from entering at will, and protect personal and property safety. However, in the current situation, after the current person is successfully identified, the following person will follow the previous person to enter and pass through, in order to evade identification, making it impossible to know the specific identity of the person evading facial recognition access control, resulting in a greatly reduced security of the access control.

[0004] Based on this, the present invention provides a face recognition access control system. Summary of the Invention

[0005] In response to the shortcomings of the existing technology, the present invention provides a face recognition access control system, which has the advantages of identifying and warning when multiple people pass by and providing prompts through a loudspeaker, thereby improving access control security and solving the problems raised in the background technology.

[0006] The present invention provides the following technical solution: a face recognition access control system, including a personnel information collection module, a data security and privacy protection module, a face and real scene detection module, a feature extraction module, a face comparison module and a system management module:

[0007] Personnel information collection module: When a person who has not entered the system needs to join the system, the person who has not entered the system needs to enter the property office in person, and then the facial recognition terminal camera in the property office will collect the facial image of the person who has not entered the system. The person who has not entered the system fills in the personal identity information by scanning the QR code generated by the QR code generator, and the identity information is associated with the facial image. In addition, when the user registers, a clear user consent process is added to ensure that the user understands and agrees to the use of their personal information;

[0008] Data security and privacy protection module: (1) Data encryption: Encrypt the transmitted and stored personnel information data, and use the SSL / TLS encryption protocol to ensure the security of data during network transmission and prevent data from being stolen or tampered with; (2) Access control: According to the responsibilities and work needs of managers, different operating permissions are assigned, such as some personnel can only view data, while some personnel can enter personnel information and set permissions, etc., to prevent data leakage; (3) Privacy policy: In the process of data use, follow the principle of minimum necessity, only collect and use necessary information related to access control management, and avoid excessive collection of user privacy data;

[0009] Face and real-scene detection module: (1) Front-end face recognition: After a person enters the face recognition access control area, the camera in front of the person automatically captures the face image without the person having to stop or make special movements; (2) Rear-end person recognition: After a person enters the face recognition access control area, a camera set at a high position continuously collects video images of the entrance and exit areas and uses algorithms and models to track and warn people in the area:

[0010] ① Initialization: Load the pre-trained target detection and tracking model;

[0011] ②Frame reading: read video frames from the camera;

[0012] ③Target detection: Use YOLOV5 algorithm to detect pedestrians;

[0013] ④Target tracking: Use DeepSORT to track pedestrians and assign IDs;

[0014] ⑤ People counting: Counting people through virtual lines or areas;

[0015] ⑥Result display: display statistical results on the video frame;

[0016] ⑦Warning sending: When the number of people passing through the virtual line at the same time exceeds the limit, a warning sound effect is played through the pygame library;

[0017] Feature extraction module: uses the MySQL database to store the feature vectors of registered faces and personnel information, and uses MySQL for fast retrieval and comparison with face recognition terminals;

[0018] Face matching module: When the user verifies the authorized "face" on the face recognition device outside the door, the access controller will unlock the electric lock and release the user after the face recognition device recognizes and confirms the legal identity, and record the entry time and the information of the passerby: When leaving the door, the user verifies the authorized "face" on the face recognition device inside the door. After the face recognition device recognizes and confirms the legal identity, the access controller will unlock the electric lock and release the user, and also record the entry time and the information of the passerby;

[0019] System management module: (1) User management: responsible for adding, deleting, modifying and other operations on system users. Different user permissions can be set. For example, administrators can perform all operations, while ordinary users can only perform basic face recognition operations. (2) Log management: records various system operations and event logs, such as user identification records, system login records, etc., for easy query and audit, so that problems can be traced back. (3) Device management: manages hardware such as cameras and access control devices in the system, including device status monitoring, parameter configuration, etc., to ensure the normal operation of the equipment.

[0020] Preferably, in addition to the door opening method of opening the electric lock to release the door, the access controller also includes a door opening method in which a valid password must be entered after sensing a valid access card to unlock the electric lock.

[0021] Preferably, the camera in front of the personnel and the camera set at a high place are both infrared cameras, and the minimum pixel requirement of the camera is 2 million pixels or above, and the minimum frame rate is 25fps-30fps.

[0022] Preferably, the camera in front of the person is installed at a distance of 1.5 meters to 1.7 meters from the ground. The installation position ensures that the face can be fully captured without obstruction, avoids backlight or direct strong light, and prevents image overexposure or shadow interference with recognition.

[0023] Preferably, the camera installation position at the height is set 2.5 meters to 2.7 meters from the ground, and it is necessary to ensure that the installation position can clearly capture the entire face recognition access control area.

[0024] Preferably, the QR code generated by the QR code generator is used to fill in the personal identity information including name, gender, detailed residential address, and ID number.

[0025] Preferably, the access controller receives verification information from the face recognition device at the entrance and the face recognition device at the exit via Wiegand signals.

[0026] Preferably, the positive pole (+) of the electric lock is connected to the positive pole (+) of the lock control output of the access controller, and the negative pole (-) of the electric lock is connected to the negative pole (-) of the lock control output of the access controller, and the power supply is used to provide appropriate voltage and current. The face recognition device includes a face recognition device at the entrance and a face recognition device at the exit.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] The facial recognition access control system of this application can promptly detect situations where multiple people attempt to force or illegally pass through the access control, which can deter potential criminals and reduce the probability of illegal intrusion incidents. In crowded places, it can avoid the congestion and pushing caused by multiple people passing through the access control at the same time, effectively prevent stampede accidents, and protect people's lives. At the same time, when safety problems arise, the system records the information of multiple people passing through and issues early warnings, which makes it easier to trace the process of the incident and provide a strong basis for subsequent investigations. Finally, for people who pass through normally, the system's early warning function can make them feel the strict security management of the venue and enhance their sense of security and comfort in the venue. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is a schematic diagram of the system modules of the present invention;

[0030] Figure 2 This is a schematic diagram of the personnel information collection module of the present invention;

[0031] Figure 3 This is a schematic diagram of the system management module of the present invention;

[0032] Figure 4 This is a schematic diagram of the data security and privacy protection module of the present invention. DETAILED DESCRIPTION

[0033] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0034] See also Figure 1-4 , a face recognition access control system, including a personnel information collection module, a data security and privacy protection module, a face and real scene detection module, a feature extraction module, a face comparison module and a system management module:

[0035] Personnel information collection module: When a person who has not entered the system needs to join the system, the person who has not entered the system needs to enter the property office in person, and then the facial recognition terminal camera in the property office will collect the facial image of the person who has not entered the system. The person who has not entered the system fills in the personal identity information by scanning the QR code generated by the QR code generator, and the identity information is associated with the facial image. In addition, when the user registers, a clear user consent process is added to ensure that the user understands and agrees to the use of their personal information;

[0036] Data security and privacy protection module: (1) Data encryption: Encrypt the transmitted and stored personnel information data, and use the SSL / TLS encryption protocol to ensure the security of data during network transmission and prevent data from being stolen or tampered with; (2) Access control: According to the responsibilities and work needs of managers, different operating permissions are assigned, such as some personnel can only view data, while some personnel can enter personnel information and set permissions, etc., to prevent data leakage; (3) Privacy policy: In the process of data use, follow the principle of minimum necessity, only collect and use necessary information related to access control management, and avoid excessive collection of user privacy data;

[0037] Face and real-scene detection module: (1) Front-end face recognition: After a person enters the face recognition access control area, the camera in front of the person automatically captures the face image without the person having to stop or make special movements; (2) Rear-end person recognition: After a person enters the face recognition access control area, a camera set at a high position continuously collects video images of the entrance and exit areas and uses algorithms and models to track and warn people in the area:

[0038] ① Initialization: Load the pre-trained target detection and tracking model;

[0039] ②Frame reading: read video frames from the camera;

[0040] ③Target detection: Use YOLOV5 algorithm to detect pedestrians;

[0041] ④Target tracking: Use DeepSORT to track pedestrians and assign IDs;

[0042] ⑤ People counting: Counting people through virtual lines or areas;

[0043] ⑥Result display: display statistical results on the video frame;

[0044] ⑦Warning sending: When the number of people passing through the virtual line at the same time exceeds the limit, a warning sound effect is played through the pygame library. The specific basic code is as follows:

[0045]

[0046]

[0047] #Open the camera

[0048] cap=cv2.VideoCapture(0)#0 indicates the default camera

[0049] if not cap.isOpened():

[0050] print("Error:Could not open camera.")

[0051] return

[0052] #Define virtual line (for counting)

[0053] line_y=300 #Assume that the virtual line is at a position where the image height is 300

[0054] counter=0

[0055] warning_threshold=2#Set the warning threshold. When the number of people passing through at the same time exceeds this value, a warning will be triggered.

[0056] warning_active=False#Warning status

[0057] while True:

[0058] ret,frame=cap.read()

[0059] if not ret:

[0060] break

[0061] #YOLOv5 detection

[0062] img=cv2.cvtColor(frame,cv2.COLOR_BGR2RGB)

[0063] img_tensor=torch.from_numpy(img).permute(2,0,1).float().div(255.0).unsqueeze(0)

[0064] with torch.no_grad():

[0065] detections=yolov5_model(img_tensor)[0]

[0066]

[0067]

[0068]

[0069] Feature extraction module: uses the MySQL database to store the feature vectors of registered faces and personnel information, and uses MySQL for fast retrieval and comparison with face recognition terminals;

[0070] Face matching module: When the user verifies the authorized "face" on the face recognition device outside the door, the access controller will unlock the electric lock and release the user after the face recognition device recognizes and confirms the legal identity, and record the entry time and the information of the passerby: When leaving the door, the user verifies the authorized "face" on the face recognition device inside the door. After the face recognition device recognizes and confirms the legal identity, the access controller will unlock the electric lock and release the user, and also record the entry time and the information of the passerby;

[0071] System management module: (1) User management: responsible for adding, deleting, modifying and other operations on system users. Different user permissions can be set. For example, administrators can perform all operations, while ordinary users can only perform basic face recognition operations; (2) Log management: records various system operations and event logs, such as user identification records, system login records, etc., for easy query and audit, so that problems can be traced; (3) Equipment management: manages hardware such as cameras and access control devices in the system, including equipment status monitoring, parameter configuration, etc., to ensure the normal operation of the equipment. This application can promptly detect situations where multiple people attempt to force or illegally pass through the access control, which can deter potential criminals and reduce the probability of illegal intrusion incidents. In crowded places, it can avoid congestion and pushing caused by multiple people passing through the access control at the same time, effectively prevent stampede accidents, and protect people's lives. At the same time, when safety problems occur, the system records the information and warnings of multiple people passing through, which is convenient for tracing the event process and providing a strong basis for subsequent investigations. Finally, for people who pass through normally, the system's warning function can make them feel the strict safety management of the venue and enhance their sense of security and comfort in the venue.

[0072] In addition to opening the electric lock to release the door, the access controller also includes a method of opening the electric lock after sensing a valid access card and entering a valid password. The camera in front of the personnel and the camera installed at a high position are both infrared cameras. The minimum pixel requirement of the camera is 2 million pixels or above, and the minimum frame rate is 25fps-30fps. The requirements of 2 million pixels or above and a minimum frame rate of 25fps-30fps can ensure the clarity of the picture and the smoothness of real-time shooting. The installation position of the camera in front of the personnel is 1.5 meters to 1.7 meters from the ground. The installation position ensures that the face can be fully captured without obstruction, avoiding backlight or direct strong light, and preventing image overexposure or shadow interference with recognition. The camera should be installed at a height of 2.5-2.7 meters from the ground. It is necessary to ensure that the installation location can clearly capture the entire face recognition access control area. The QR code generated by the QR code generator is used to fill in the personnel identity information including name, gender, detailed residential address, and ID number. The access control controller receives the verification information of the face recognition device at the entrance and the face recognition device at the exit through the Wiegand signal. The positive pole (+) of the electric lock is connected to the positive pole (+) of the lock control output of the access control controller, and the negative pole (-) of the electric lock is connected to the negative pole (-) of the lock control output of the access control controller. The power supply provides appropriate voltage and current. The face recognition device includes the face recognition device at the entrance and the face recognition device at the exit.

[0073] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0074] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A face recognition access control system, characterized in that: It includes personnel information collection module, data security and privacy protection module, face and real scene detection module, feature extraction module, face comparison module and system management module: Personnel information collection module: When a person who has not entered the system needs to join the system, the person who has not entered the system needs to enter the property office in person, and then the facial recognition terminal camera in the property office will collect the facial image of the person who has not entered the system. The person who has not entered the system fills in the personal identity information by scanning the QR code generated by the QR code generator, and the identity information is associated with the facial image. In addition, when the user registers, a clear user consent process is added to ensure that the user understands and agrees to the use of their personal information; Data security and privacy protection module: (1) Data encryption: Encrypt the transmitted and stored personnel information data, and use the SSL / TLS encryption protocol to ensure the security of data during network transmission and prevent data from being stolen or tampered with; (2) Access control: According to the responsibilities and work needs of managers, different operating permissions are assigned, such as some personnel can only view data, while some personnel can enter personnel information and set permissions, etc., to prevent data leakage; (3) Privacy policy: In the process of data use, follow the principle of minimum necessity, only collect and use necessary information related to access control management, and avoid excessive collection of user privacy data; Face and real-scene detection module: (1) Front-end face recognition: After a person enters the face recognition access control area, the camera in front of the person automatically captures the face image without the person having to stop or make special movements; (2) Rear-end person recognition: After a person enters the face recognition access control area, a camera set at a high position continuously collects video images of the entrance and exit areas and uses algorithms and models to track and warn people in the area: ① Initialization: Load the pre-trained target detection and tracking model; ②Frame reading: read video frames from the camera; ③Target detection: Use YOLOV5 algorithm to detect pedestrians; ④Target tracking: Use DeepSORT to track pedestrians and assign IDs; ⑤ People counting: Counting people through virtual lines or areas; ⑥Result display: display statistical results on the video frame; ⑦Warning sending: When the number of people passing through the virtual line at the same time exceeds the limit, a warning sound effect is played through the pygame library; Feature extraction module: uses the MySQL database to store the feature vectors of registered faces and personnel information, and uses MySQL for fast retrieval and comparison with face recognition terminals; Face matching module: When the user verifies the authorized "face" on the face recognition device outside the door, the access controller will unlock the electric lock and release the user after the face recognition device recognizes and confirms the legal identity. It also records the entry time and the information of the passerby. When leaving the door, the user verifies the authorized "face" on the face recognition device inside the door. After the face recognition device recognizes and confirms the legal identity, the access controller will unlock the electric lock and release the user. It also records the entry time and the information of the passerby. System management module: (1) User management: responsible for adding, deleting, modifying and other operations on system users. Different user permissions can be set. For example, administrators can perform all operations, while ordinary users can only perform basic face recognition operations. (2) Log management: Record various system operations and event logs, such as user identification records, system login records, etc., to facilitate query and audit, so that problems can be traced back; (3) Equipment management: Manage hardware such as cameras and access control devices in the system, including equipment status monitoring, parameter configuration, etc., to ensure the normal operation of the equipment.

2. A facial recognition access control system according to claim 1, characterized in that: In addition to the door opening method of opening the electric lock to release the door, the access controller also includes a door opening method in which a valid password must be entered after sensing a valid access card to open the electric lock.

3. The facial recognition access control system according to claim 1, characterized in that: The camera in front of the personnel and the camera installed at a high place are both infrared cameras. The minimum pixel requirement of the camera is 2 million pixels or above, and the minimum frame rate is 25fps-30fps.

4. The facial recognition access control system according to claim 1, characterized in that: The camera in front of the person is installed at a distance of 1.5 meters to 1.7 meters from the ground.

5. The face recognition access control system according to claim 1, characterized in that: The camera installation position at the height is set at 2.5 meters to 2.7 meters from the ground.

6. The face recognition access control system according to claim 1, characterized in that: The QR code generated by the QR code generator is used to fill in the personal identity information including name, gender, detailed residential address, and ID number.

7. The face recognition access control system according to claim 1, characterized in that: The access controller receives verification information from the face recognition device at the entrance and the face recognition device at the exit through the Wiegand signal.

8. The face recognition access control system according to claim 1, characterized in that: The positive pole (+) of the electric lock is connected to the positive pole (+) of the lock control output of the access controller, and the negative pole (-) of the electric lock is connected to the negative pole (-) of the lock control output of the access controller. The power supply is used to provide appropriate voltage and current. The face recognition device includes an entry face recognition device and an exit face recognition device.

Images