Community remote monitoring video transmission method and system

Abstract

The invention relates to the technical field of video monitoring processing, and particularly discloses a community remote monitoring video transmission method and system, and the method comprises the following steps: carrying out the framing of a monitoring video in an edge node, obtaining a video frame, and obtaining the movement track of a person based on the video frame; obtaining a track sequence, grouping the personnel based on the track sequence, and determining a target track corresponding to the group; generating an image set, determining a target video, and transmitting the image set and the target video to a data center; and obtaining an image set A where the to-be-determined image is located, and extracting a target video corresponding to the image set A for display. According to the invention, the operation pressure of the monitoring system can be reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of video surveillance processing, and particularly relates to a method and system for community remote surveillance video transmission. Background Art

[0002] With the continuous progress of technology, the application of remote surveillance in community management has become increasingly popular. By installing high-definition cameras, sensors and other intelligent devices in the community, managers can real-time grasp the dynamic situation of public areas, such as key areas like entrances and exits, corridors, parking lots, etc., so as to quickly capture abnormal behaviors, and through linkage with intelligent devices such as alarm systems and access control systems, achieve automatic early warning and remote control, and timely handle various security hazards.

[0003] With the wide application of surveillance devices in the community, the number of devices and image clarity have been continuously improved. A large number of high-definition cameras and advanced sensors are installed in various areas, resulting in a sharp increase in real-time transmitted video data. The continuously rising data volume poses a greater pressure on network transmission bandwidth, leading to a significant decrease in upload speed. At the same time, the convergence of a large amount of data also poses a severe test to the storage and processing capabilities of the central server, further increasing the burden of system operation. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for community remote surveillance video transmission to solve the following technical problems:

[0005] With the wide application of surveillance devices in the community, the number of devices and image clarity have been continuously improved. A large number of high-definition cameras and advanced sensors are installed in various areas, resulting in a sharp increase in real-time transmitted video data. The continuously rising data volume poses a greater pressure on network transmission bandwidth, leading to a significant decrease in upload speed. At the same time, the convergence of a large amount of data also poses a severe test to the storage and processing capabilities of the central server, further increasing the burden of system operation.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A method for community remote surveillance video transmission includes the following steps:

[0008] Obtain the surveillance video of a single surveillance device, transmit the surveillance video to a preset edge node, perform frame division on the surveillance video in the edge node to obtain video frames, obtain the position of a single person in a single video frame, and connect the positions in the order of the time axis to obtain the movement trajectory of the person corresponding to the single surveillance device;

[0009] Sort the movement trajectories in the order of the time axis, denoted as trajectory sorting, group the personnel based on the trajectory sorting, and determine the target trajectories corresponding to the groups;

[0010] Mark the personnel corresponding to a single target trajectory as target personnel, obtain the face images of the target personnel, generate an image set CJH=(C1, C2,..., Cm), where Cm represents the face image of the m-th target personnel, denote all the surveillance videos corresponding to the target trajectory as target videos, and transmit the image set and the target videos to the data center together;

[0011] Obtain the face images of the preset surveillance personnel, denoted as undetermined images, obtain the image set A where the undetermined images are located, and the data center extracts and displays the target videos corresponding to the image set A.

[0012] As a further solution of the present invention: determining the target trajectories corresponding to the groups includes:

[0013] Generate undetermined trajectories based on the trajectory sorting. When the similarity degree between any two undetermined trajectories is greater than a preset similarity degree threshold, divide them into the same group, mark the undetermined trajectories in the same group as candidate trajectories, and the total similarity degree between the target trajectories and the remaining candidate trajectories is the largest.

[0014] As a further solution of the present invention: generating undetermined trajectories based on the trajectory sorting includes:

[0015] Obtain any two adjacent movement trajectories in the trajectory sorting, and mark them as movement trajectory i and movement trajectory i + 1 respectively. The sorting position of movement trajectory i in the trajectory sorting is to the left of the sorting position of movement trajectory i + 1 in the trajectory sorting;

[0016] Connect the end point of movement trajectory i and the start point of movement trajectory i + 1 with a straight line to obtain a new movement trajectory to replace movement trajectory i and movement trajectory i + 1;

[0017] Obtain a new trajectory sorting, repeat the above steps until there is only one movement trajectory in a certain trajectory sorting, and denote this movement trajectory as an undetermined trajectory.

[0018] As a further solution of the present invention: obtaining the similarity degree of the undetermined trajectories includes:

[0019] Obtain the trajectory sortings corresponding to undetermined trajectory j and undetermined trajectory k, and mark them as trajectory sorting J1 and trajectory sorting K1 respectively. Obtain the movement trajectories at the y-th position in trajectory sorting J1 and trajectory sorting K1, and denote them as sub-trajectory J1y and sub-trajectory K1y respectively;

[0020] Generate the curve corresponding to the sub-trajectory J1y in the preset space rectangular coordinate system, obtain the tangent slope of each point on the curve, sort the tangent slopes in ascending order according to the magnitude of the target distance, and obtain the slope sorting P1. The target distance represents the distance between the point on the tangent corresponding to the tangent slope and the starting point, and the starting point represents the point on the curve corresponding to the starting point of the sub-trajectory;

[0021] Obtain the slope sorting P2 corresponding to the sub-trajectory K1y, and calculate the similarity between the sub-trajectory J1y and the sub-trajectory K1y:

[0022]

[0023] where B represents the total number of tangent slopes in the slope sorting, and P1b and P2b respectively represent the tangent slope at the b-th position in the slope sorting P1 and the tangent slope at the b-th position in the slope sorting P2;

[0024] Calculate the total similarity as the degree of similarity.

[0025] As a further solution of the present invention: in the process of grouping the personnel, it further includes:

[0026] Calculate the first difference DY = P1tot - P2tot, where P1tot and P2tot respectively represent the total number of the tangent slopes in the slope sorting P1 and the slope sorting P2;

[0027] When the first difference is greater than the preset first difference threshold, mark the first difference as an abnormal difference, count the total number of the abnormal differences, and when the total number is greater than the preset total number threshold, do not divide the pending trajectory j and the pending trajectory k into the same group.

[0028] As a further solution of the present invention: in the process of grouping the personnel, it further includes:

[0029] Calculate the second difference DR = TOTj - TOTk, where TOTj and TOTk respectively represent the total number of the moving trajectories corresponding to the pending trajectory j and the pending trajectory k;

[0030] When the second difference is greater than the preset second difference threshold, do not divide the pending trajectory j and the pending trajectory k into the same group.

[0031] As a further solution of the present invention: after jointly transmitting the image set and the target video to the data center, it further includes:

[0032] The data center deletes the image set and the target video whose total storage duration reaches the preset duration.

[0033] A community remote monitoring video transmission system, characterized by comprising:

[0034] Edge transmission module: Obtain the monitoring video of a single monitoring device, transmit the monitoring video to a preset edge node, perform frame segmentation on the monitoring video in the edge node to obtain video frames, obtain the position of a single person in a single video frame, and connect the positions in chronological order to obtain the movement trajectory of the person corresponding to the single monitoring device;

[0035] Central transmission module: Sort the movement trajectories in chronological order, denoted as trajectory sorting, group the persons based on the trajectory sorting, and determine the target trajectory corresponding to the group;

[0036] Mark the person corresponding to a single target trajectory as a target person, obtain the face image of the target person, generate an image set CJH=(C1, C2,..., Cm), where Cm represents the face image of the mth target person, denote all the monitoring videos corresponding to the target trajectory as the target video, and jointly transmit the image set and the target video to the data center;

[0037] Display module: Obtain the face image of a preset monitoring person, denoted as the pending image, obtain the image set A where the pending image is located, and the data center extracts and displays the target video corresponding to the image set A.

[0038] Advantages of the present invention: Compared with the prior art:

[0039] 1) In the present invention, frame segmentation, trajectory extraction, and grouping are first performed at the edge node, and only the video segments corresponding to the finally determined target trajectory and the face images of the target persons will be uploaded, rather than all the original video data being uploaded to the server, which effectively reduces the bandwidth occupied in the data transmission link on the network.

[0040] 2) By grouping the trajectories of multiple persons and using a target trajectory to represent the passing data of the persons in the same group, only the necessary target videos and face images are uploaded, avoiding the occupation of server resources by duplicate or redundant monitoring data, thereby extending the effective working life of the server, reducing the server retrieval and analysis delay caused by excessive data volume, and improving the overall operation efficiency of the system;

[0041] 3) In actual monitoring applications, it is often necessary to check the activity trajectories or appearance time periods of specific individuals. In the traditional method, it is necessary to "search carpet-style" for the images of the target person in a large number of video files, which is not only time-consuming and laborious, but also prone to false detection or missed detection. In the present invention, by grouping and annotating the trajectories of personnel at the edge node, the central side only needs to determine the group to which the query person belongs based on the face (pending image) of the query person, and then quickly lock the corresponding target video for playback and analysis. This greatly shortens the query path, especially in a community environment with a large number of monitored personnel and complex scene environments, the retrieval speed will be greatly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The present invention will be further described below with reference to the accompanying drawings.

[0043] Figure 1 is a schematic flowchart of a method for transmitting community remote monitoring videos according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts shall fall within the protection scope of the present invention.

[0045] Please refer to Figure 1 As shown, the present invention is a method for transmitting community remote monitoring videos, including the following steps:

[0046] Obtain the monitoring video of a single monitoring device, transmit the monitoring video to a preset edge node, perform frame division on the monitoring video in the edge node to obtain video frames, obtain the position of a single person in a single video frame, and connect the positions in the order of the time axis to obtain the movement trajectory of the person corresponding to the single monitoring device.

[0047] It should be noted that in the present invention, the real-time video stream of a single monitoring device can be obtained through the RTSP protocol. After decoding the streaming media data using a video decoder, the video data is stably transmitted to a preset edge node through network transmission protocols such as TCP / IP or UDP. After reaching the edge node, frame splitting is performed using the cv2.VideoCapture module in the OpenCV library. This method can read the video frame by frame at a set frame rate, thereby decomposing the continuous video stream into a series of static images. Each frame image provides an independent data unit for subsequent processing. After obtaining each video frame, object detection algorithms based on deep learning, such as YOLO or SSD, are used to detect the people in the image, and the position information of a single person in this frame is accurately extracted through image segmentation and positioning techniques. Subsequently, by introducing object tracking algorithms, such as the Kalman filter or optical flow tracking, the positions of the same people detected in consecutive frames are matched and associated. With the help of the time stamps of the frames, the position information of the people in each frame is connected in sequence along the time axis, thus forming the complete movement trajectory of the person, and finally realizing the trajectory extraction of the person corresponding to a single monitoring device;

[0048] Sort the movement trajectories in the order of the time axis, denoted as trajectory sorting. Group the people based on the trajectory sorting, and determine the target trajectory corresponding to the group;

[0049] In a preferred embodiment of the present invention, determining the target trajectory corresponding to the group includes:

[0050] Generate candidate trajectories based on the trajectory sorting. When the similarity degree between any two candidate trajectories is greater than a preset similarity degree threshold, divide them into the same group, and mark the candidate trajectories in the same group as candidate trajectories for selection. The total similarity degree between the target trajectory and the remaining candidate trajectories for selection is the largest.

[0051] It is worth noting that candidate trajectories are generated according to the recorded trajectory sorting, and the similarity degree between the candidate trajectories is evaluated. When the similarity degree between any two candidate trajectories exceeds the preset threshold, they are divided into the same group, and all candidate trajectories within the group are marked as candidate trajectories for selection. Subsequently, by comparing the total similarity degree between the candidate trajectories for selection in each group, the candidate trajectory with the largest total similarity degree is selected as the target trajectory. This can ensure that the selected target trajectory best represents the movement characteristics of all people within the group, thereby further improving the accuracy and efficiency of subsequent video transmission and processing;

[0052] In a preferred case of this embodiment, generating candidate trajectories based on the trajectory sorting includes:

[0053] Obtain any two adjacent moving trajectories in the trajectory sorting, and respectively label them as moving trajectory i and moving trajectory i+1. The sorting position of moving trajectory i in the trajectory sorting is to the left of the sorting position of moving trajectory i+1 in the trajectory sorting;

[0054] Connect the end point of the moving trajectory i and the start point of the moving trajectory i+1 by a straight line to obtain a new moving trajectory to replace the moving trajectory i and the moving trajectory i+1;

[0055] Obtain a new trajectory sorting, and repeat the above steps until there is only one moving trajectory in a certain trajectory sorting, and denote this moving trajectory as the pending trajectory;

[0056] It should be noted that first, by traversing the sorted trajectory data, adjacent two moving trajectories are obtained in turn and are respectively labeled as moving trajectory i and moving trajectory i+1, where moving trajectory i is before moving trajectory i+1 in the data list; then, by using the coordinates of the end point of moving trajectory i and the start point of moving trajectory i+1 as the two end points of a straight line, a simple straight line equation is used to generate a new straight line segment as the fusion result of these two trajectories, so as to replace the original two trajectories, and the updated trajectory data is sorted again. This process is realized through iterative loops until there is only one moving trajectory left in the sorted data, which is denoted as the pending trajectory;

[0057] In another preferred case of this embodiment, obtaining the similarity degree of the pending trajectory includes:

[0058] Obtain the corresponding trajectory sortings of the pending trajectory j and the pending trajectory k, and respectively label them as trajectory sorting J1 and trajectory sorting K1. Obtain the moving trajectories at the y-th position in the trajectory sortings J1 and K1, and respectively denote them as sub-trajectory J1y and sub-trajectory K1y;

[0059] Generate a curve corresponding to the sub-trajectory J1y in a preset space rectangular coordinate system, obtain the tangent slope of each point on the curve, sort the tangent slopes in ascending order according to the size of the target distance, and obtain a slope sorting P1. The target distance represents the distance between the point on the tangent corresponding to the tangent slope and the starting point, and the starting point represents the point on the curve corresponding to the starting point of the sub-trajectory;

[0060] Obtain the slope sorting P2 corresponding to the sub-trajectory K1y, and calculate the similarity between the sub-trajectory J1y and the sub-trajectory K1y:

[0061]

[0062] Among them, B represents the total number of tangent slopes in the slope sorting, and P1b and P2b respectively represent the tangent slope at the b-th position in the slope sorting P1 and the tangent slope at the b-th position in the slope sorting P2;

[0063] Calculate the total similarity as the degree of similarity;

[0064] It can be understood that, for a fine evaluation of the similarity between the to-be-determined trajectories, first, the corresponding sequences of the to-be-determined trajectory j and the to-be-determined trajectory k in the original trajectory sorting are respectively obtained, and the y-th moving trajectory is extracted from each sequence as the sub-trajectory J1y and the sub-trajectory K1y. Then, curve fitting is performed on each sub-trajectory in a preset three-dimensional Cartesian coordinate system to generate a curve, and the tangent slope of each point on the curve is calculated using the difference method or other numerical methods. Then, an ascending order sorting is performed according to the distance from the starting point corresponding to each tangent slope to this point, obtaining the slope sortings P1 and P2; finally, according to the total number B of tangent slopes in the slope sorting and the difference between the tangent slopes of each corresponding position, the similarity between the two sub-trajectories is calculated, and the similarities of each sub-trajectory are accumulated to obtain the total similarity degree; this method realizes the effective fusion of the trajectories within the group and the accurate selection of the target trajectory through the merging of continuous trajectories and the accurate quantification of similarity, thereby improving the efficiency of subsequent video transmission and processing, reducing redundant data, and optimizing the overall operation performance of the system;

[0065] In a preferred case of this embodiment, during the process of grouping the personnel, it further includes:

[0066] Calculate the first difference DY = P1tot - P2tot, where P1tot and P2tot respectively represent the total number of the tangent slopes in the slope sorting P1 and the slope sorting P2;

[0067] When the first difference is greater than a preset first difference threshold, mark the first difference as an abnormal difference, count the total number of the abnormal differences, and when the total number is greater than a preset total number threshold, do not divide the to-be-determined trajectory j and the to-be-determined trajectory k into the same group.

[0068] In another preferred case of this embodiment, during the process of grouping the personnel, it further includes:

[0069] Calculate the second difference DR = TOTj - TOTk, where TOTj and TOTk respectively represent the total number of the moving trajectories corresponding to the to-be-determined trajectory j and the to-be-determined trajectory k;

[0070] When the second difference is greater than a preset second difference threshold, do not divide the to-be-determined trajectory j and the to-be-determined trajectory k into the same group.

[0071] Mark the person corresponding to the single target trajectory as the target person, obtain the face image of the target person, generate an image set CJH = (C1, C2,..., Cm), where Cm represents the face image of the m-th target person, record all the surveillance videos corresponding to the target trajectory as the target videos, and jointly transmit the image set and the target videos to the data center;

[0072] It can be understood that according to the determined target trajectory, the corresponding person is identified and marked at the edge node to ensure that the system can accurately distinguish the target person. Using the pre-trained person recognition algorithm, the person matching the target trajectory can be quickly screened out from a large amount of person data; subsequently, the face image of the target person is extracted from the real-time video frame through the built-in face detection module (for example, using the detection algorithm based on convolutional neural network), and the image is pre-processed such as normalization and image enhancement to generate a clear face image sequence, which are successively composed into an image set; at the same time, by performing timestamp synchronization processing on the video data in the period covered by the target trajectory, all the surveillance videos in this period are integrated and marked as the target videos to ensure the accurate correspondence between the video content and the person trajectory. Finally, using data compression and encryption technologies, the generated image set and the target videos are packaged and transmitted to the data center to achieve the secure storage and efficient invocation of the data; not only ensuring the accurate confirmation of the target person's identity and the precise matching of the key video data, but also greatly reducing the redundancy of the transmitted data and improving the efficiency of subsequent retrieval and analysis of the system;

[0073] Obtain the face image of the preset surveillance person, denoted as the pending image, obtain the image set A where the pending image is located, and the data center extracts and displays the target videos corresponding to the image set A;

[0074] In a preferred embodiment of the present invention, after jointly transmitting the image set and the target videos to the data center, it further includes:

[0075] The data center deletes the image set and the target videos whose total storage duration reaches the preset duration.

[0076] A community remote surveillance video transmission system includes:

[0077] Edge transmission module: Obtain the surveillance video of a single surveillance device, transmit the surveillance video to a preset edge node, perform frame splitting on the surveillance video in the edge node to obtain video frames, obtain the position of a single person in a single video frame, and connect the positions in the order of the time axis to obtain the movement trajectory of the person corresponding to the single surveillance device;

[0078] Central transmission module: Sort the movement trajectories in the order of the time axis, denoted as trajectory sorting. Group the personnel based on the trajectory sorting and determine the target trajectories corresponding to the groups.

[0079] Mark the personnel corresponding to a single target trajectory as target personnel, obtain the face images of the target personnel, generate an image set CJH=(C1, C2,..., Cm), where Cm represents the face image of the m-th target personnel. Denote all the surveillance videos corresponding to the target trajectory as target videos, and transmit the image set and the target videos to the data center together.

[0080] Display module: Obtain the face images of the preset surveillance personnel, denoted as pending images, obtain the image set A where the pending images are located, and the data center extracts the target videos corresponding to the image set A for display.

[0081] The above has described a specific embodiment of the present invention in detail, but the content is only the preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the present invention application shall still fall within the scope covered by the patent of the present invention.

Claims

1. A method for community remote monitoring video transmission, characterized in that, It includes the following steps: Obtain the monitoring video of a single monitoring device, transmit the monitoring video to a preset edge node, perform frame splitting on the monitoring video in the edge node to obtain video frames, obtain the position of a single person in a single video frame, and connect the positions in the order of the time axis to obtain the movement trajectory of the person corresponding to the single monitoring device; Sort the movement trajectories in the order of the time axis, denoted as trajectory sorting, group the persons based on the trajectory sorting, and determine the target trajectory corresponding to the group; Mark the persons corresponding to a single target trajectory as target persons, obtain the face images of the target persons, generate an image set CJH=(C1, C2,..., Cm), where Cm represents the face image of the m-th target person, denote all the monitoring videos corresponding to the target trajectory as target videos, and transmit the image set and the target videos to the data center together; Obtain the face image of a preset monitored person, denoted as the to-be-determined image, obtain the image set A where the to-be-determined image is located, and the data center extracts and displays the target videos corresponding to the image set A.

2. The method for community remote monitoring video transmission according to claim 1, characterized in that Determining the target trajectory corresponding to the group includes: Generate to-be-determined trajectories based on the trajectory sorting. When the similarity degree between any two to-be-determined trajectories is greater than a preset similarity degree threshold, divide them into the same group, mark the to-be-determined trajectories in the same group as candidate trajectories, and the total similarity degree between the target trajectory and the remaining candidate trajectories is the largest.

3. The method for community remote monitoring video transmission according to claim 2, characterized in that Generating to-be-determined trajectories based on the trajectory sorting includes: Obtain any two adjacent movement trajectories in the trajectory sorting, and mark them as movement trajectory i and movement trajectory i+1 respectively. The sorting position of movement trajectory i in the trajectory sorting is to the left of the sorting position of movement trajectory i+1 in the trajectory sorting; Connect the end point of movement trajectory i and the start point of movement trajectory i+1 with a straight line to obtain a new movement trajectory to replace movement trajectory i and movement trajectory i+1; Obtain a new trajectory sorting, repeat the above steps until there is only one movement trajectory in a certain trajectory sorting, and denote this movement trajectory as the to-be-determined trajectory.

4. A method for community remote monitoring video transmission according to claim 2, characterized in that, Obtaining the similarity degree of the to-be-determined trajectory includes: Obtain the trajectory sortings corresponding to to-be-determined trajectory j and to-be-determined trajectory k, and mark them as trajectory sorting J1 and trajectory sorting K1 respectively. Obtain the movement trajectory at the y-th position in trajectory sorting J1 and trajectory sorting K1, and denote them as sub-trajectory J1y and sub-trajectory K1y respectively; Generate a curve corresponding to sub-trajectory J1y in a preset space rectangular coordinate system, obtain the tangent slope of each point on the curve, sort the tangent slopes in ascending order according to the size of the target distance, and obtain a slope sorting P1. The target distance represents the distance between the point on the tangent corresponding to the tangent slope and the starting point, and the starting point represents the point on the curve corresponding to the starting point of the sub-trajectory; Obtain the slope sorting P2 corresponding to sub-trajectory K1y, and calculate the similarity between sub-trajectory J1y and sub-trajectory K1y: Among them, B represents the total number of tangent slopes in the slope sorting. P1b and P2b respectively represent the tangent slope at the b-th position in the slope sorting P1 and the tangent slope at the b-th position in the slope sorting P2; Calculate the total similarity as the degree of similarity.

5. A method for community remote monitoring video transmission according to claim 4, characterized in that, During the process of grouping the personnel, it further includes: Calculate the first difference DY = P1tot - P2tot, where P1tot and P2tot respectively represent the total number of the tangent slopes in the slope sorting P1 and the slope sorting P2; When the first difference is greater than a preset first difference threshold, mark the first difference as an abnormal difference, count the total number of the abnormal differences. When the total number is greater than a preset total number threshold, do not divide the pending trajectory j and the pending trajectory k into the same group.

6. The method for community remote monitoring video transmission according to claim 4, characterized in that During the process of grouping the personnel, it further includes: Calculate the second difference DR = TOTj - TOTk, where TOTj and TOTk respectively represent the total number of the movement trajectories corresponding to the pending trajectory j and the pending trajectory k; When the second difference is greater than a preset second difference threshold, do not divide the pending trajectory j and the pending trajectory k into the same group.

7. A method for community remote monitoring video transmission according to claim 1, characterized in that, After jointly transmitting the image set and the target video to the data center, it further includes: The data center deletes the image set and the target video whose stored total duration reaches the preset duration.

8. A community remote monitoring video transmission system, characterized in that It includes: Edge transmission module: Obtain the monitoring video of a single monitoring device, transmit the monitoring video to a preset edge node, perform frame division on the monitoring video in the edge node to obtain video frames, obtain the position of a single person in a single video frame, and connect the positions in the order of the time axis to obtain the movement trajectory of the person corresponding to the single monitoring device; Central transmission module: Sort the movement trajectories in the order of the time axis, denoted as trajectory sorting, group the personnel based on the trajectory sorting, and determine the target trajectory corresponding to the group; Mark the person corresponding to a single target trajectory as a target person, obtain the face image of the target person, generate an image set CJH = (C1, C2,..., Cm), where Cm represents the face image of the m-th target person, denote all the monitoring videos corresponding to the target trajectory as the target video, and jointly transmit the image set and the target video to the data center; Display module: Obtain the face image of a preset monitored person, denoted as the pending image, obtain the image set A where the pending image is located, and the data center extracts the target video corresponding to the image set A for display.

Images