Cross-camera face tracking method and device based on spatio-temporal correlation, and medium

Abstract

This invention discloses a cross-camera face tracking method based on spatiotemporal correlation. By adopting the idea of ​​face tracking, this method effectively reduces the number of face recognition operations, thereby reducing the time and resource overhead of face feature extraction and face comparison in a massive face database. Utilizing the relatively fixed position of a face relative to the overall pedestrian population, it achieves face tracking based on a single camera and realizes cross-camera face tracking based on the spatiotemporal correlation information of a face appearing in the field of view of each camera. Simultaneously, by uniformly processing the identity information of pedestrians on the same trajectory, it improves the accuracy of face recognition.

Description

Technical Field

[0001] This invention relates to the field of computer vision technology, and more specifically, to a method, device, and medium for cross-camera face tracking based on spatiotemporal correlation. Background Technology

[0002] Facial recognition technology has developed rapidly and is now widely used in identity verification. However, deploying specific frontal face capture devices for identity verification in high-traffic areas (such as subway stations) often leads to congestion, reduced throughput, and security risks. The method of automatically capturing facial images using video sensors for identity recognition results in unstable image quality and frequent facial feature comparisons, further reducing recognition accuracy and efficiency. Simultaneously, recognizing each facial image independently can lead to the same face being identified as belonging to different individuals, further decreasing recognition accuracy.

[0003] Current cross-camera pedestrian tracking methods based on overlapping fields of view suffer from the problem of pedestrians being lost due to the excessively large overlapping field of view. Furthermore, the application scenarios of these methods are limited; tracking is impossible when the camera's shooting areas are not connected. On the other hand, target tracking methods that do not rely on overlapping fields of view typically use pedestrian re-identification technology. This involves determining the location of the same pedestrian in different cameras before tracking. However, these methods require identification from a large number of cameras, resulting in low tracking efficiency. Summary of the Invention

[0004] To address at least one deficiency or improvement need in the prior art, this invention provides a cross-camera face tracking method based on spatiotemporal correlation, which solves the technical problem of how to reduce the waste of computing resources and time overhead caused by face recognition frame by frame in real-time video streams.

[0005] To achieve the above objectives, in a first aspect, the present invention provides a cross-camera face tracking method based on spatiotemporal correlation, comprising the following steps:

[0006] Based on the physical location information of the cameras and the preset walking speed of the pedestrians, the spatiotemporal correlation information of the pedestrians appearing in the field of view of different cameras is initialized.

[0007] The system tracks pedestrians within the field of view of a single camera, selects facial regions based on the tracked pedestrian positions, and uploads the facial regions for facial recognition at fixed time intervals.

[0008] When a pedestrian disappears from the field of view of a certain camera, the field of view of another camera at a physical location adjacent to that camera is queried based on the spatiotemporal correlation information, and the spatiotemporal correlation information is dynamically updated based on the query results.

[0009] Merge all trajectory information of the same pedestrian, retain only all recognition results obtained by facial recognition of the face region uploaded during the tracking phase, and use the recognition result with the highest similarity as the pedestrian's identity information.

[0010] Furthermore, the initialization of the spatiotemporal correlation information of a pedestrian appearing in the field of view of different cameras based on the physical location information of the cameras and the preset walking speed of the pedestrian specifically includes:

[0011] Assign a unique camera number to each camera based on the physical location distribution map of the cameras, and mark the shortest walkable distance between the cameras.

[0012] Using the camera as the vertex and the shortest walkable path distance to the relevant camera as the edge, a directed weighted graph is constructed between the cameras. Then, combined with the pedestrian's preset walking speed, the spatiotemporal correlation information of the pedestrian appearing in the field of view of different cameras is initialized.

[0013] Furthermore, the step of tracking pedestrians within the field of view of a single camera, selecting a face region based on the tracked pedestrian position, and uploading the face region for face recognition at fixed time intervals specifically includes:

[0014] Track pedestrians within the field of view of a single camera, recording pedestrian characteristics and the start and end times of their appearance in that field of view;

[0015] Based on the pedestrian location obtained from tracking, a face region is selected, and the face region is divided and uploaded at fixed time intervals for face detection and recognition.

[0016] Furthermore, the tracking of pedestrians within the field of view of a single camera, recording pedestrian characteristics and the start and end times of their appearance in that field of view, specifically includes:

[0017] A single camera tracks and detects all pedestrians passing by from the starting point of its field of view, assigns a unique pedestrian number to each pedestrian in the field of view, records the start time of the current tracking, and then uses a Kalman filter to predict the position of the pedestrian in the next frame. When the Intersection over Union (IOU) between the predicted position of a pedestrian and the detection result is less than a set threshold, the tracking and detection of that pedestrian in the field of view is stopped, and the current tracking stop time, the camera number of the current camera, and the pedestrian features are recorded.

[0018] Furthermore, the step of selecting a face region based on the tracked pedestrian location, dividing and uploading the face region at fixed time intervals for face detection and recognition specifically includes:

[0019] When tracking pedestrians in real time, a rectangular area of ​​a preset proportion at the current pedestrian position is taken as the face region at preset time intervals, and the position information of the pedestrian in the original image to which the face region belongs is saved as the face region information.

[0020] Upload the original image and the corresponding face region information, then use the location information in the face region information as a face bounding box for face detection and recognition, and then save the obtained identity information and corresponding similarity in the pedestrian trajectory information.

[0021] Furthermore, when the pedestrian disappears from the field of view of a certain camera, the field of view of another camera physically located adjacent to that camera is queried based on spatiotemporal correlation information, and the spatiotemporal correlation information is dynamically updated based on the query results, specifically including:

[0022] When a pedestrian disappears from the field of view of a certain camera, the field of view of another camera at a physical location adjacent to that camera is queried based on spatiotemporal correlation information to perform pedestrian re-identification.

[0023] Record the time interval between pedestrian appearances between cameras at adjacent physical locations over a period of time, and use statistical methods to dynamically update the spatiotemporal correlation information of the camera's field of view based on the query results.

[0024] Furthermore, when a pedestrian disappears from the field of view of a certain camera, the process of querying the field of view of another camera physically located adjacent to that camera based on spatiotemporal correlation information to perform pedestrian re-identification specifically includes:

[0025] Obtain pedestrian tracking information, and based on the tracking stop time, query the starting time of tracking for cameras at other adjacent physical locations that match the fixed time interval according to spatiotemporal correlation information to perform continuous cross-camera face tracking.

[0026] Furthermore, the step of recording the time interval between pedestrian appearances at cameras at adjacent physical locations over a period of time, and dynamically updating the spatiotemporal correlation information of the camera's field of view using statistical methods based on the query results, specifically includes:

[0027] Based on the acquired pedestrian tracking information, after querying the camera where the pedestrian will be located in the next stage according to the spatiotemporal correlation information, the time interval and correlation between the two cameras that the pedestrian appears in are recorded. When the set update time threshold is reached, a statistical method is used to dynamically update the spatiotemporal correlation information of the camera's field of view using the range of the most frequent occurrences.

[0028] In a second aspect, the present invention provides an electronic device including at least one processing unit and at least one storage unit, wherein the storage unit stores a computer program that, when executed by the processing unit, enables the processing unit to perform the steps of any of the methods described above.

[0029] Thirdly, the present invention provides a storage medium storing a computer program executable by an access authentication device, which, when run on the access authentication device, enables the access authentication device to perform the steps of any of the methods described above.

[0030] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects:

[0031] This invention discloses a cross-camera face tracking method based on spatiotemporal correlation. By adopting the idea of ​​face tracking, this method effectively reduces the number of face recognition operations, thereby reducing the time and resource overhead of face feature extraction and face comparison in a massive face database. Utilizing the relatively fixed position of a face relative to the overall pedestrian population, it achieves face tracking based on a single camera and realizes cross-camera face tracking based on the spatiotemporal correlation information of a face appearing in the field of view of each camera. Simultaneously, by uniformly processing the identity information of pedestrians on the same trajectory, it improves the accuracy of face recognition. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 A flowchart illustrating a cross-camera face tracking method based on spatiotemporal correlation provided in an embodiment of the present invention;

[0034] Figure 2 This is a schematic diagram of a scene of a directed weighted graph between cameras constructed according to an embodiment of the present invention;

[0035] Figure 3 A block diagram illustrating an electronic device suitable for implementing the methods described above, provided in an embodiment of the present invention. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0037] The terms "comprising" or "having," and any variations thereof, in the specification, claims, or accompanying drawings of this application are intended to cover a non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the steps or units listed, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such processes, methods, products, or apparatus.

[0038] Facial recognition technology has developed rapidly and is now widely used in identity verification. However, deploying specific frontal face capture devices for identity verification in high-traffic areas (such as subway stations) often leads to congestion, reduced throughput, and security risks. The method of automatically capturing facial images using video sensors for identity recognition results in unstable image quality and frequent facial feature comparisons, further reducing recognition accuracy and efficiency. Simultaneously, recognizing each facial image independently can lead to the same face being identified as belonging to different individuals, resulting in decreased recognition accuracy.

[0039] Face tracking is a feasible method to solve the above problems. Face tracking involves locating target faces in image sequences in real time. By constructing the motion trajectory of each pedestrian, the number of face recognition operations is reduced. At the same time, the recognition result with the highest similarity to a face database in a trajectory is used as the pedestrian's identity information, thereby ultimately improving the accuracy and efficiency of face recognition.

[0040] like Figure 1 As shown, in a specific embodiment of a subway station scenario, a cross-camera face tracking method based on spatiotemporal correlation mainly includes steps S1-S4.

[0041] S1. Based on the physical location information of the cameras in the subway station, initialize the spatiotemporal correlation information of pedestrians appearing in the field of view of different cameras.

[0042] More specifically, based on the construction drawings of the subway station, an abstract planar positional diagram of the cameras is drawn, each camera is assigned a unique number (CID), and the distance (X) between the cameras is marked. i,j (X) i,jThis represents the shortest distance a pedestrian can travel from camera i to camera j via a walkable path, not the straight-line distance between the camera installation locations. For example... Figure 2 As shown, the shortest walkable path distance X between camera 4 and camera 6 is... 4,6 It should be 1+3=4, not 1+3=4. Then, taking the camera as a vertex, and using the shortest walkable path distance X to the relevant camera... i,j A directed weighted graph G(V,E) between cameras is constructed for the edges, and the spatiotemporal correlation information of pedestrians appearing in the field of view of different cameras is initialized based on the preset walking speed of pedestrians of 0.5 m / s.

[0043] S2. Track pedestrians in the field of view of a single camera, select the face region based on the tracked pedestrian position, and upload the face region for face recognition at fixed time intervals.

[0044] More specifically, step S2 includes the following steps S21-S22:

[0045] S21. Track pedestrians using a single camera and record pedestrian characteristics and the start and end times when a pedestrian appears in a field of view.

[0046] Furthermore, a single camera tracks and detects all pedestrians passing by from the starting point of its field of view, and assigns a unique PID to each pedestrian in that field of view. i Record the start time Ts of the current tracking. i Then, a Kalman filter is used to predict the position of the pedestrian in the next frame. When the Intersection over Union (IOU) between the predicted and detected positions of a pedestrian is less than a set threshold, the tracking and detection of that pedestrian in the field of view is stopped, and the current tracking stop time Te is recorded. i Current camera ID (CID) j Pedestrian characteristics PF i Then cancel the pedestrian's PID. i .

[0047] S22. Select the face region from the pedestrian location obtained by tracking, divide and upload the face region at fixed time intervals for detection and recognition.

[0048] Furthermore, during real-time pedestrian tracking, every 1 second at preset intervals, the upper third of the rectangular area at the current pedestrian's position is designated as the face region. The position information (X1, Y1, X2, Y2) of the PID to which this face region belongs in the original image is saved as a face region information. Simultaneously with uploading the original image, the camera also uploads all face region information from that image. The position information within these face region information is then used as a pre-selected face bounding box. After face detection and recognition, the obtained identity information and corresponding similarity scores are saved in the PID's trajectory information.

[0049] S3, numbered CID i After a pedestrian disappears from the field of view of a camera, the system queries the CID based on spatiotemporal correlation information. i The field of view of another camera in an adjacent physical location is used to dynamically update the spatiotemporal correlation information based on the query results.

[0050] More specifically, step S3 includes the following steps S31-S32:

[0051] S31: When a pedestrian disappears from the field of view of a certain camera, the field of view of another camera at a physical location adjacent to that camera is queried based on the spatiotemporal correlation information, thereby achieving rapid pedestrian re-identification.

[0052] Furthermore, when the camera CID j Uploaded pedestrian PID i After the tracking information is sent to the central controller, the central controller queries other cameras at adjacent physical locations based on the tracking end time and spatiotemporal correlation information to find cameras that match the tracking start time at a fixed time interval, and then performs continuous face tracking across cameras.

[0053] S32: Record the time interval between pedestrian appearances between cameras at adjacent physical locations over a period of time, and use statistical methods to dynamically update the spatiotemporal correlation information of the camera's field of view based on the query results.

[0054] Furthermore, regarding camera CID j Uploaded pedestrian PID i Based on the tracking information and spatiotemporal correlation information, the camera CID of the pedestrian in the next stage can be queried. k Then, record the pedestrian's PID. i The time interval between the two cameras mentioned above is recorded, and when the set update time threshold is reached, a statistical method is used to dynamically update the spatiotemporal correlation information of the camera's field of view using the range of the most frequent time intervals.

[0055] S4. Merge all trajectories of the same pedestrian with the results of uploaded facial region recognition, and use the identity with the highest similarity as the identity of the pedestrian.

[0056] More specifically, the main controller is based on the pedestrian PID i All received trajectory information is merged, and only all recognition results obtained from the face region uploaded during the tracking phase are retained. The recognition result with the highest similarity is used as the identity of the pedestrian.

[0057] Figure 3 A block diagram schematically illustrates an electronic device suitable for implementing the methods described above, according to an embodiment of the present invention. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of the present invention.

[0058] like Figure 3 As shown, the electronic device 1000 described in this embodiment includes a processor 1001, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage portion 1008 into a random access memory (RAM) 1003. The processor 1001 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 1001 may also include onboard memory for caching purposes. The processor 1001 may include a single processing unit or multiple processing units for performing different actions of the method flow according to embodiments of this disclosure.

[0059] RAM 1003 stores various programs and data required for the operation of system 1000. Processor 1001, ROM 1002, and RAM 1003 are interconnected via bus 1004. Processor 1001 performs various operations of the method flow according to embodiments of the present disclosure by executing programs in ROM 1002 and / or RAM 1003. It should be noted that the programs may also be stored in one or more memories other than ROM 1002 and RAM 1003. Processor 1001 may also perform various operations of the method flow according to embodiments of the present disclosure by executing programs stored in said one or more memories.

[0060] According to embodiments of this disclosure, the electronic device 1000 may further include an input / output (I / O) interface 1005, which is also connected to a bus 1004. The system 1000 may also include one or more of the following components connected to the I / O interface 1005: an input section 1006 including a keyboard, mouse, etc.; an output section 1007 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1008 including a hard disk, etc.; and a communication section 1009 including a network interface card such as a LAN card, modem, etc. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the I / O interface 1005 as needed. A removable medium 1011, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 1010 as needed so that computer programs read from it can be installed into the storage section 1008 as needed.

[0061] The method flow according to embodiments of this disclosure can be implemented as a computer software program. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable storage medium, the computer program containing program code for performing the methods shown in the flowchart. In such embodiments, the computer program can be downloaded and installed from a network via communication section 1009, and / or installed from removable medium 1011. When the computer program is executed by processor 1001, it performs the functions defined in the system of embodiments of this disclosure. According to embodiments of this disclosure, the systems, devices, apparatuses, modules, or units described above can be implemented using computer program modules.

[0062] Embodiments of the present invention also provide a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of the present disclosure.

[0063] According to embodiments of this disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, such as including but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM). The storage medium may be a ROM, an optical storage device, a magnetic storage device, or any suitable combination thereof. In embodiments of this disclosure, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used or combined with an instruction execution system, apparatus, or device. For example, according to embodiments of this disclosure, the computer-readable storage medium may include one or more memories other than the ROM 1002 and / or RAM 1003 described above.

[0064] It should be noted that the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product.

[0065] The flowcharts or block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. Furthermore, it should be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0066] Those skilled in the art will understand that the features described in the various embodiments and / or claims of this disclosure can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in this disclosure. In particular, the features described in the various embodiments and / or claims of this disclosure can be combined and / or combined in various ways without departing from the spirit and teachings of this disclosure, and all such combinations and / or combinations fall within the scope of this disclosure.

[0067] Although this disclosure has been shown and described with reference to specific exemplary embodiments thereof, those skilled in the art will understand that various changes in form and detail may be made to this disclosure without departing from the spirit and scope of the disclosure as defined by the appended claims and their equivalents. Therefore, the scope of this disclosure should not be limited to the above embodiments, but should be defined not only by the appended claims, but also by their equivalents.

Claims

1. A cross-camera face tracking method based on spatiotemporal correlation, characterized in that, Including the following steps: Based on the physical location information of the cameras and the preset walking speed of the pedestrians, the spatiotemporal correlation information of the pedestrians appearing in the field of view of different cameras is initialized. The system tracks pedestrians within the field of view of a single camera, selects facial regions based on the tracked pedestrian positions, and uploads the facial regions for facial recognition at fixed time intervals. When a pedestrian disappears from the field of view of a certain camera, the field of view of another camera at a physical location adjacent to that camera is queried based on the spatiotemporal correlation information, and the spatiotemporal correlation information is dynamically updated based on the query results. The dynamic update includes recording the time interval between pedestrian appearances between adjacent cameras over a period of time, and using statistical methods to dynamically update the spatiotemporal correlation information using the range of time intervals with the highest frequency of appearance. Merge all trajectory information of the same pedestrian, retain only all recognition results obtained by facial recognition of the face region uploaded during the tracking phase, and use the recognition result with the highest similarity as the pedestrian's identity information; The process of tracking pedestrians within the field of view of a single camera, selecting facial regions based on the tracked pedestrian positions, and uploading the facial regions for face recognition at fixed time intervals specifically includes: Track pedestrians within the field of view of a single camera, recording pedestrian characteristics and the start and end times of their appearance in that field of view; Based on the pedestrian location obtained from tracking, the face region is selected, and the face region is divided and uploaded at fixed time intervals for face detection and recognition; The step of selecting a face region based on the pedestrian's location obtained through tracking, and dividing and uploading the face region at fixed time intervals for face detection and recognition specifically includes: When tracking pedestrians in real time, a rectangular area of ​​a preset proportion at the current pedestrian position is taken as the face region at preset time intervals, and the position information of the pedestrian in the original image to which the face region belongs is saved as the face region information. Upload the original image and the corresponding face region information, then use the location information in the face region information as a face bounding box for face detection and recognition, and then save the obtained identity information and corresponding similarity in the pedestrian trajectory information.

2. The face tracking method as described in claim 1, characterized in that, The initialization of the spatiotemporal correlation information of a pedestrian appearing in the field of view of different cameras based on the physical location information of the cameras and the preset walking speed of the pedestrian specifically includes: Assign a unique camera number to each camera based on the physical location distribution map of the cameras, and mark the shortest walkable distance between the cameras. Using the camera as the vertex and the shortest walkable path distance to the relevant camera as the edge, a directed weighted graph is constructed between the cameras. Then, combined with the pedestrian's preset walking speed, the spatiotemporal correlation information of the pedestrian appearing in the field of view of different cameras is initialized.

3. The face tracking method as described in claim 1, characterized in that, The tracking of pedestrians within the field of view of a single camera, recording pedestrian characteristics and the start and end times of their appearance in that field of view, specifically includes: A single camera tracks and detects all pedestrians passing by from the starting point of its field of view, assigns a unique pedestrian number to each pedestrian in the field of view, records the start time of the current tracking, and then uses a Kalman filter to predict the position of the pedestrian in the next frame. When the Intersection over Union (IOU) between the predicted position of a pedestrian and the detection result is less than a set threshold, the tracking and detection of that pedestrian in the field of view is stopped, and the current tracking stop time, the camera number of the current camera, and the pedestrian features are recorded.

4. The face tracking method as described in claim 1, characterized in that, When a pedestrian disappears from the field of view of a certain camera, the field of view of another camera at a physical location adjacent to that camera is queried based on spatiotemporal correlation information, and the spatiotemporal correlation information is dynamically updated based on the query results. Specifically, this includes: When a pedestrian disappears from the field of view of a certain camera, the field of view of another camera at a physical location adjacent to that camera is queried based on spatiotemporal correlation information to perform pedestrian re-identification. Record the time interval between pedestrian appearances between cameras at adjacent physical locations over a period of time, and use statistical methods to dynamically update the spatiotemporal correlation information of the camera's field of view based on the query results.

5. The face tracking method as described in claim 4, characterized in that, When a pedestrian disappears from the field of view of a certain camera, the field of view of another camera at a physical location adjacent to that camera is queried based on spatiotemporal correlation information to perform pedestrian re-identification. Specifically, this includes: Obtain pedestrian tracking information, and based on the tracking stop time, query the starting time of tracking for cameras at other adjacent physical locations that match the fixed time interval according to spatiotemporal correlation information to perform continuous cross-camera face tracking.

6. The face tracking method as described in claim 4, characterized in that, The process of recording the time interval between pedestrian appearances between cameras at adjacent physical locations over a period of time, and dynamically updating the spatiotemporal correlation information of the camera's field of view using statistical methods based on the query results, specifically includes: Based on the acquired pedestrian tracking information, after querying the camera where the pedestrian will be located in the next stage according to the spatiotemporal correlation information, the time interval and correlation between the two cameras where the pedestrian appears are recorded. When the set update time threshold is reached, a statistical method is used to dynamically update the spatiotemporal correlation information of the camera's field of view using the range of the most frequent occurrences.

7. An electronic device, characterized in that, It includes at least one processing unit and at least one storage unit, wherein the storage unit stores a computer program that, when executed by the processing unit, causes the processing unit to perform the steps of the method according to any one of claims 1 to 6.

8. A storage medium, characterized in that, It stores a computer program executable by an access authentication device, which, when run on the access authentication device, causes the access authentication device to perform the steps of the method according to any one of claims 1 to 6.

Images