Passenger track identification method and system

Abstract

The invention provides a passenger track identification method and system, and belongs to the technical field of target identification, and the method comprises the steps: constructing a monitoring camera network topology structure, dividing regions according to a people flow density principle, and calculating the weight of a vertex of each monitoring camera based on a queuing theory model; according to the calculated vertex weight, performing vertex weighting operation on the constructed monitoring camera network topology structure, and constructing a vertex weighted directed graph; the weights are distributed to the vertexes in the vertex weighted directed graph, an improved shortest path algorithm is combined and used, the weights between the vertexes, the historical path weight and the path length are comprehensively considered, a comprehensive evaluation function is constructed, an optimal video stream path is selected, and a target passenger is determined according to the optimal video stream path. According to the method, the queuing theory model and the vertex weighted directed graph theory are applied to realize intelligent selection of monitoring camera videos, so that the video processing efficiency and the resource utilization rate are improved, and the efficiency and the accuracy of passenger behavior track restoration are improved.

Description

Technical Field

[0001] The present invention relates to the field of target recognition technology, and in particular to a method and system for cross-camera passenger trajectory recognition based on video technology. Background Art

[0002] Most of the existing passenger trajectory recognition methods are based on road network topology search and enumeration of all cameras to restore passenger trajectories. However, in the implementation process, they rely too much on image analysis algorithms and ignore the possible behavioral trajectories of passengers during the ride, which leads to low search efficiency and poor search accuracy. Summary of the invention

[0003] The object of the present invention is to provide a method and system for identifying passenger trajectories using a camera based on video technology to solve at least one technical problem existing in the above-mentioned background technology.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] In a first aspect, the present invention provides a passenger trajectory identification method, characterized by comprising:

[0006] Build the surveillance camera network topology;

[0007] Based on the constructed surveillance camera network topology, the area is divided according to the principle of human flow density, and the weight of each surveillance camera vertex is calculated based on the queuing theory model;

[0008] According to the calculated vertex weights, a vertex weighting operation is performed on the constructed surveillance camera network topology structure to construct a vertex weighted directed graph;

[0009] Weights are assigned to vertices in a vertex weighted directed graph. An improved shortest path algorithm is used in combination with the weights between vertices, historical path weights, and path lengths to construct a comprehensive evaluation function and select the optimal video stream path. The target passenger is determined based on the optimal video stream path.

[0010] Furthermore, a surveillance camera network topology structure is constructed, including: collecting location information and network connection information of surveillance cameras, including historical data of human traffic and historical vertex weights; and calculating a comprehensive evaluation value of the directed edge direction of the surveillance camera network topology structure based on the human traffic information in the historical data.

[0011] Furthermore, based on the pedestrian flow information in the historical data, the pedestrian flow weight and historical vertex weight of each directed edge are calculated, and the pedestrian flow weight and historical vertex weight are multiplied by the corresponding weight coefficients respectively, and then added together to obtain a comprehensive evaluation value of the directed edge direction.

[0012] Furthermore, the calculation of the passenger flow weight of each directed edge includes: the total passenger flow in the station in a certain time period is k, and the passenger flow from one node to another node in the period is K, then the passenger flow weight of the corresponding directed edge is

[0013] Furthermore, the calculation of the historical vertex weight of each directed edge includes: the vertex in the weighted directed graph represents a camera in the monitoring area, and a vertex is connected to multiple directed edges. Let the directed edges connected to vertex V be E1, E2, ... E m , and the corresponding traffic weights are w p1 、w p2 …w pm ; The historical vertex weight is j∈(1,2,3,...,m).

[0014] Furthermore, the improved shortest path algorithm is the comprehensive evaluation function, which is to track the optimal path of the passenger's movement trajectory and find the target passenger according to the optimal path;

[0015] Comprehensive evaluation function of the optimal path for target passengers:

[0016] To determine the shortest path for the target passenger to reach a certain facility node, the shortest path specifically includes the vertex to reach a certain facility node. The path corresponding to this vertex is defined as the shortest path for the passenger to reach this node, denoted as min(L i )=min{W(v)+L(v)+P(v)}, where W(v) represents the weight value of the vertex; L(v) represents the weight value of the historical path, and the calculation formula is In the surveillance video data within a certain period of time, n i Indicates that the target passenger passes through path L i N represents the number of times all target passengers have passed through all historical paths; P(v) represents the length of the path from the starting node to the node, calculated according to the Euclidean distance formula. Suppose the starting node is (x1, y1), and the vertices under this node are (x i ,y i ),but min(L)={(L1),9L2),...(L n )}, min(L) is the optimal path to track the target passenger’s movement trajectory.

[0017] In a second aspect, the present invention provides a passenger trajectory identification system, comprising:

[0018] The first building module is used to build a monitoring camera network topology structure;

[0019] The calculation module is used to divide the area according to the principle of human flow density based on the constructed surveillance camera network topology structure, and calculate the weight of each surveillance camera vertex based on the queuing theory model;

[0020] The second construction module is used to perform vertex weighting operations on the constructed surveillance camera network topology structure according to the calculated vertex weights to construct a vertex weighted directed graph;

[0021] The screening module is used to assign weights to vertices in the vertex weighted directed graph, combine the improved shortest path algorithm, comprehensively consider the weights between each vertex, historical path weights and path lengths, construct a comprehensive evaluation function, and screen the optimal video stream path; determine the target passenger based on the optimal video stream path.

[0022] In a third aspect, the present invention provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium is used to store computer instructions, and when the computer instructions are executed by a processor, the passenger trajectory identification method as described in the first aspect is implemented.

[0023] In a fourth aspect, the present invention provides a computer device comprising a memory and a processor, wherein the processor and the memory communicate with each other, the memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the passenger trajectory identification method as described in the first aspect.

[0024] In a fifth aspect, the present invention provides an electronic device, comprising: a processor, a memory and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory so that the electronic device executes instructions for implementing the passenger trajectory identification method as described in the first aspect.

[0025] The beneficial effects of the present invention are as follows: by applying the queuing theory model and vertex weighted directed graph theory, intelligent selection of surveillance camera videos is achieved, video processing efficiency and resource utilization are improved, and the efficiency and accuracy of passenger behavior trajectory restoration are improved.

[0026] Additional advantages of the present invention will be more clearly given in the following description or learned through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0028] Figure 1 This is a flow chart of a method for cross-camera passenger trajectory recognition based on video technology according to an embodiment of the present invention.

[0029] Figure 2 A schematic diagram of a vertex weighted directed graph constructed according to an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The embodiments described below by the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention.

[0031] It should be understood by those skilled in the art that unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those skilled in the art to which this invention belongs.

[0032] It should also be understood that terms, such as those defined in commonly used dictionaries, should be understood to have a meaning consistent with that in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless as defined herein.

[0033] Those skilled in the art will appreciate that, unless otherwise stated, the singular forms "a", "an", "said" and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements and / or groups thereof.

[0034] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. Different embodiments or examples described in this specification and features of different embodiments or examples may be combined and combined by those skilled in the art without contradiction.

[0035] To facilitate understanding of the present invention, the present invention is further explained below with reference to specific embodiments in conjunction with the accompanying drawings, and the specific embodiments do not constitute a limitation on the embodiments of the present invention.

[0036] Those skilled in the art should understand that the drawings are merely schematic diagrams of embodiments, and the components in the drawings are not necessarily necessary for implementing the present invention.

[0037] Most of the existing passenger trajectory restoration methods use traversal search to search all camera videos or topological network search and repeated search, which leads to the problem of low efficiency and poor accuracy in personnel tracking. In order to solve the above problems, the present invention proposes a method for efficiently selecting surveillance camera videos based on queuing theory and vertex weighted directed graph theory. By assigning weights to cameras and using weighted calculations, the method can find the optimal motion path of target passengers at urban rail transit nodes, thereby achieving efficient tracking of the action trajectory of target passengers across cameras. Drawing on the principles of queuing theory, the camera is used as a vertex, and a weighted directed graph is constructed. The optimal monitoring path is determined by calculating the weights. Considering the connection and weights between different cameras, the surveillance camera video can be efficiently selected, avoiding the inefficiency of traversal search and repeated search. At the same time, it combines the optimal motion path of target passengers at urban rail transit nodes with personnel tracking through the idea of ​​vertex weighted directed graphs, thereby improving accuracy and efficiency.

[0038] Example 1

[0039] In this embodiment 1, a passenger trajectory identification system is first provided, including: a first construction module, which is used to construct a surveillance camera network topology structure; a calculation module, which is used to divide the area according to the principle of human flow density based on the constructed surveillance camera network topology structure, and calculate the weight of each surveillance camera vertex based on the queuing theory model; a second construction module, which is used to perform vertex weighting operations on the constructed surveillance camera network topology structure according to the calculated vertex weights, and construct a vertex weighted directed graph; a screening module, which is used to assign weights to vertices in the vertex weighted directed graph, and use an improved shortest path algorithm in combination, comprehensively consider the weights between each vertex, the historical path weights and the path length, construct a comprehensive evaluation function, screen the optimal video stream path, and determine the target passenger according to the optimal video stream path.

[0040] In this embodiment, a passenger trajectory identification method is implemented using the above-mentioned system, including: constructing a surveillance camera network topology structure; dividing the area according to the principle of crowd density based on the constructed surveillance camera network topology structure, and calculating the weight of each surveillance camera vertex based on a queuing theory model; performing vertex weighting operations on the constructed surveillance camera network topology structure according to the calculated vertex weights to construct a vertex weighted directed graph; assigning weights to vertices in the vertex weighted directed graph, combining the use of an improved shortest path algorithm, comprehensively considering the weights between each vertex, historical path weights and path lengths, constructing a comprehensive evaluation function, and selecting the optimal video stream path; processing and analyzing the selected surveillance camera video to achieve real-time monitoring and early warning.

[0041] In this embodiment, constructing a surveillance camera network topology structure includes: collecting location information and network connection information of the surveillance camera, including historical data of human traffic and historical vertex weights; and calculating a comprehensive evaluation value of the directed edge direction of the surveillance camera network topology structure according to the human traffic information in the historical data. According to the human traffic information in the historical data, the human traffic weight and the historical vertex weight of each directed edge are calculated, and the human traffic weight and the historical vertex weight are multiplied by the corresponding weight coefficients, and then added to obtain a comprehensive evaluation value of the directed edge direction.

[0042] Based on the video data in the subway station, the passenger flow data information of a certain period of time is counted. Let the total passenger flow in the station be k, and calculate and count the passenger flow from one node to another, for example, from the entrance v 11 To the security checkpoint 21 The flow of people is K, then the flow weight of the directed edge is

[0043] The vertex in the weighted directed graph represents a camera in the monitoring area. A vertex is connected to multiple directed edges. Let the directed edges connected to vertex V be E1, E 2, …E m , and the corresponding traffic weights are w p1 、w p2 …w pm The historical vertex weight is

[0044] The proposed improved shortest path algorithm is equivalent to the proposed comprehensive evaluation function, that is, the optimal path to track the passenger's movement trajectory and to find the target passenger based on the optimal path.

[0045] The calculation formula of the comprehensive evaluation function of the optimal path for target passengers is:

[0046] To determine the shortest path for the target passenger to reach a certain facility node, the shortest path specifically includes the vertex to reach a certain facility node. The path corresponding to this vertex is defined as the shortest path for the passenger to reach this node, denoted as min(L i )=min{W(v)+L(v)+P(v)}, where W(v) represents the weight value of the vertex; L(v) represents the weight value of the historical path, and the calculation formula is In the surveillance video data within a certain period of time, n i Indicates that the target passenger passes through path L i N represents the number of times all target passengers have passed through all historical paths; P(v) represents the length of the path from the starting node to the node, calculated according to the Euclidean distance formula. Suppose the starting node is (x1, y1), and the vertices under this node are (xi ,y i ),but min(L)={(L1),(L2),...(L n )}, min(L) is the optimal path to track the target passenger’s movement trajectory.

[0047] Example 2

[0048] like Figure 1 As shown, in this embodiment 2, a cross-camera passenger trajectory recognition method based on video technology is provided. This method does not restore the passenger trajectory by traversing either the camera or the road network topology, but rather realizes intelligent selection of surveillance camera videos by applying queuing theory models and vertex weighted directed graph theory to improve video processing efficiency and resource utilization, thereby accelerating the efficiency and accuracy of passenger behavior trajectory restoration.

[0049] The cross-camera passenger trajectory recognition method based on video technology described in this embodiment includes the following process steps:

[0050] Step 1: Construct the network topology of surveillance cameras: Collect the location information and network connection information of surveillance cameras, including historical data of pedestrian flow and historical vertex weights. According to the pedestrian flow information in the historical data, calculate the pedestrian flow weight and historical vertex weight of each directed edge. After the calculation is completed, adjust the parameters according to the actual operating environment and follow the formula: Directed edge direction weight = α* pedestrian flow weight + β* historical vertex weight. Multiply the pedestrian flow weight and historical vertex weight by the corresponding weight coefficient and add them together to obtain the comprehensive evaluation value of the directed edge direction.

[0051] Specifically, based on the video data in the subway station, the passenger flow data information of a certain period of time is counted. Let the total passenger flow in the station be k, and calculate and count the passenger flow from one node to another node, for example, from the entrance v 11 To the security checkpoint 21 The flow of people is K, then the flow weight of the directed edge is

[0052] The vertex in the weighted directed graph represents a camera in the monitoring area. A vertex is connected to multiple directed edges. Let the directed edges connected to vertex V be E1, E 2, …E m , and the corresponding traffic weights are w p1 、w p2 …w pm The historical vertex weight is

[0053] Step 2: Determine monitoring requirements and real-time scene information: Analyze actual monitoring requirements and determine the monitoring areas, focus points, and priorities that need to be monitored. Collect real-time scene information, such as crowd density, abnormal event prediction, etc., for subsequent monitoring decisions and path selection.

[0054] Step 3: Calculate vertex weights: Divide the area according to the principle of crowd density, and calculate the weight of each surveillance camera vertex based on the queuing theory model. When calculating the camera vertex weight, factors such as the camera's processing capability, queue length, and response time should be considered comprehensively.

[0055] Specifically, the passenger flow density is divided into areas: first, the subway station is divided into areas: such as entrance and exit areas, platform areas, staircase areas, etc., and the area of ​​each area is calculated as S. The video surveillance data of a certain time period is collected, and the number of people in the time period is calculated as A. Then the passenger flow density in the area is The weighted average of the density calculated in each area is recorded as the initial passenger flow density D0. If D>D0, it is divided into a high-density area.

[0056] Based on the queuing theory model, the model is mapped to the cross-camera passenger trajectory recognition analysis, and the surveillance camera video analysis process is regarded as a queuing system, the surveillance cameras available for analysis are regarded as "customers", and the analysis capability of the surveillance camera is regarded as "service capability". The queuing theory model includes factors such as customer arrival rate, service rate capability, and queue time, which are mapped to the surveillance camera video analysis. When calculating the camera vertex weight, factors such as the camera's processing capability, queue length, and response time are considered. The formula for calculating the surveillance camera vertex weight is as follows:

[0057] Assume that the number of surveillance cameras in the surveillance camera network topology is n, and the surveillance camera vertex weight set is W = {w 11 , w 12 ,…,w ij},(where w ij is the vertex v ij weight).

[0058] w ij =FPS ij ·T ij DPI ij

[0059] FPS ij Indicates the number of frames processed by the camera per second. The specific calculation method is: Where F represents the total number of frames in the target passenger video, and T represents the video duration. ij Indicates the time from when the target passenger enters the camera's field of view to when he leaves the surveillance area. DPI ijIndicates the resolution of the surveillance camera.

[0060] Step 4: Construct a vertex weighted directed graph: Figure 2 As shown, according to the calculated vertex weights, the directed graph is subjected to vertex weighting operations. Weights are assigned to the vertices in the directed graph, reflecting the processing capabilities and resource utilization of different cameras.

[0061] Step 5: Optimal path selection: When searching for a path, the optimal path is selected by comprehensively considering the node weight, historical path weight, and path length. An improved shortest path algorithm is used to comprehensively consider the weights and path lengths between nodes, and to perform a comprehensive evaluation in combination with the historical path weights. A comprehensive evaluation function is defined to weight the node weights, historical path weights, and path lengths with appropriate weight coefficients to select a video stream path with the best performance and resource utilization.

[0062] Specifically, in this embodiment, the proposed improved shortest path algorithm is equivalent to the proposed comprehensive evaluation function, that is, the optimal path for tracking the passenger's movement trajectory, and the target passenger is found according to the optimal path.

[0063] The calculation formula of the comprehensive evaluation function of the optimal path for target passengers is:

[0064] To determine the shortest path for the target passenger to reach a certain facility node, the shortest path specifically includes the vertex to reach a certain facility node. The path corresponding to this vertex is defined as the shortest path for the passenger to reach this node, denoted as min(L i )=min{W(v)+L(v)+P(v)}, where W(v) represents the weight value of the vertex; L(v) represents the weight value of the historical path, and the calculation formula is In the surveillance video data within a certain period of time, n i Indicates that the target passenger passes through path L i N represents the number of times all target passengers have passed through all historical paths; P(v) represents the length of the path from the starting node to the node, calculated according to the Euclidean distance formula. Suppose the starting node is (x1, y1), and the vertices under this node are (x i ,y i ),but min(L)={(L1),(L2),...(L n )}, min(L) is the optimal path to track the target passenger’s movement trajectory.

[0065] In this embodiment, the selected surveillance camera video is processed using target recognition, motion detection, behavior analysis and other technologies to achieve real-time monitoring and early warning functions. Specifically, target recognition, motion detection, behavior analysis and the like all use existing technologies, which will not be repeated here, and are briefly described as follows: the video stream data is continuously acquired by the surveillance camera, and the collected video frames are preprocessed, including noise reduction, image scaling, normalization and other operations. Passenger target features are extracted based on machine vision and machine learning algorithms, and target recognition and detection are performed based on the trained target recognition model, and the identified target information (such as passenger location, identity information) is recorded and stored. Motion detection technology is used to analyze the pixel changes between adjacent frames in the video sequence to detect whether the target passenger is moving. Based on the above operations, the target behavior in the video stream is understood and analyzed, and the target behavior category is judged by analyzing and synthesizing the target's motion trajectory, posture changes and other information. Finally, the above results of target recognition, motion detection and behavior analysis can be displayed on the monitoring screen to identify the target passenger and obtain the target passenger's motion trajectory.

[0066] Example 3

[0067] This embodiment 3 provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by a processor, the passenger trajectory identification method as described above is implemented. The method includes:

[0068] Build the surveillance camera network topology;

[0069] Based on the constructed surveillance camera network topology, the area is divided according to the principle of human flow density, and the weight of each surveillance camera vertex is calculated based on the queuing theory model;

[0070] According to the calculated vertex weights, a vertex weighting operation is performed on the constructed surveillance camera network topology structure to construct a vertex weighted directed graph;

[0071] Weights are assigned to vertices in a vertex weighted directed graph. An improved shortest path algorithm is used in combination with the weights between vertices, historical path weights, and path lengths to construct a comprehensive evaluation function, select the optimal video stream path, and determine the target passenger based on the optimal video stream path.

[0072] Example 4

[0073] This embodiment 4 provides a computer device, including a memory and a processor, the processor and the memory communicate with each other, the memory stores program instructions executable by the processor, the processor calls the program instructions to execute the above-mentioned passenger trajectory identification method, the method comprising:

[0074] Build the surveillance camera network topology;

[0075] Based on the constructed surveillance camera network topology, the area is divided according to the principle of human flow density, and the weight of each surveillance camera vertex is calculated based on the queuing theory model;

[0076] According to the calculated vertex weights, a vertex weighting operation is performed on the constructed surveillance camera network topology structure to construct a vertex weighted directed graph;

[0077] Weights are assigned to vertices in a vertex weighted directed graph. An improved shortest path algorithm is used in combination with the weights between vertices, historical path weights, and path lengths to construct a comprehensive evaluation function, select the optimal video stream path, and determine the target passenger based on the optimal video stream path.

[0078] Example 5

[0079] This embodiment 5 provides an electronic device, including: a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory, so that the electronic device executes instructions for implementing the above-mentioned passenger trajectory identification method, the method comprising:

[0080] Build the surveillance camera network topology;

[0081] Based on the constructed surveillance camera network topology, the area is divided according to the principle of human flow density, and the weight of each surveillance camera vertex is calculated based on the queuing theory model;

[0082] According to the calculated vertex weights, a vertex weighting operation is performed on the constructed surveillance camera network topology structure to construct a vertex weighted directed graph;

[0083] Weights are assigned to vertices in a vertex weighted directed graph. An improved shortest path algorithm is used in combination with the weights between vertices, historical path weights, and path lengths to construct a comprehensive evaluation function, select the optimal video stream path, and determine the target passenger based on the optimal video stream path.

[0084] In summary, the cross-camera passenger trajectory identification method based on video technology described in the embodiment of the present invention can intelligently select surveillance camera videos and optimize the path selection of video streams through the application of queuing theory and vertex weighted directed graph theory. In this way, the problem of resource waste existing in traditional monitoring systems can be avoided, system resources can be utilized to the maximum extent, and resource utilization can be improved. By modeling the network topology of surveillance cameras and constructing vertex weighted directed graphs, the present invention can select a video stream path with the best performance according to the performance indicators and queue conditions of the cameras. In this way, the time delay of video processing can be reduced and the processing efficiency of the monitoring system can be improved. According to the real-time scene and monitoring requirements, the surveillance camera video is intelligently selected, and real-time processing and analysis are performed. By optimizing the selection and processing of video streams, a faster and real-time monitoring solution can be provided to enhance the real-time performance and reliability of the system. By determining the monitoring requirements and real-time scene information, personalized monitoring customization can be performed according to different monitoring target areas, focus points and priorities. In this way, the user's monitoring needs can be met more accurately and customized monitoring services can be provided. Based on queuing theory and vertex weighted directed graph theory, the optimal path and video stream are selected by algorithm to achieve intelligent monitoring decision-making. This can improve the intelligence level of the monitoring system, reduce human intervention, and reduce the impact of human errors and subjective factors on monitoring decisions.

[0085] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0086] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0087] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0088] These computer program instructions can also be loaded onto a computer or other programmable data processing device, and a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0089] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative work on the basis of the technical solution disclosed in the present invention should be included in the scope of protection of the present invention.

Claims

1. A passenger trajectory identification method, characterized in that: include: Build the surveillance camera network topology; Based on the constructed surveillance camera network topology, the area is divided according to the principle of human flow density, and the weight of each surveillance camera vertex is calculated based on the queuing theory model; According to the calculated vertex weights, a vertex weighting operation is performed on the constructed surveillance camera network topology structure to construct a vertex weighted directed graph; Weights are assigned to vertices in a vertex weighted directed graph. An improved shortest path algorithm is used in combination with the weights between vertices, historical path weights, and path lengths to construct a comprehensive evaluation function, select the optimal video stream path, and determine the target passenger based on the optimal video stream path.

2. The passenger trajectory identification method according to claim 1, characterized in that: Constructing a surveillance camera network topology structure includes: collecting location information and network connection information of surveillance cameras, including historical data of human traffic and historical vertex weights; and calculating a comprehensive evaluation value of the directed edge direction of the surveillance camera network topology structure based on the human traffic information in the historical data.

3. The passenger trajectory identification method according to claim 2, characterized in that: According to the pedestrian flow information in the historical data, the pedestrian flow weight and historical vertex weight of each directed edge are calculated, and the pedestrian flow weight and historical vertex weight are multiplied by the corresponding weight coefficient respectively, and then added together to obtain the comprehensive evaluation value of the directed edge direction.

4. The passenger trajectory identification method according to claim 3, characterized in that: The calculation of the passenger flow weight of each directed edge includes: the total passenger flow in the station during a certain period of time is k, and the passenger flow from one node to another node during this period is K, then the passenger flow weight of the corresponding directed edge is 5. The passenger trajectory identification method according to claim 3, characterized in that: Calculating the historical vertex weight of each directed edge includes: The vertex in the weighted directed graph represents a camera in the monitoring area. A vertex is connected to multiple directed edges. Let the directed edges connected to vertex V be E1, E 2, …E m , and the corresponding traffic weights are w p1 、w p2 …w pm ; The historical vertex weight is 6. The passenger trajectory identification method according to claim 1, characterized in that: The improved shortest path algorithm is a comprehensive evaluation function, which is to track the optimal path of the passenger's movement trajectory and find the target passenger according to the optimal path; Comprehensive evaluation function of the optimal path for target passengers: To determine the shortest path for the target passenger to reach a certain facility node, the shortest path specifically includes the vertex to reach a certain facility node. The path corresponding to this vertex is defined as the shortest path for the passenger to reach this node, denoted as min(L i )=min{W(v)+L(v)+P(v)}, where W(v) represents the weight value of the vertex; L(v) represents the weight value of the historical path, and the calculation formula is In the surveillance video data within a certain period of time, n i Indicates that the target passenger passes through path L i N represents the number of times all target passengers have passed through all historical paths; P(v) represents the length of the path from the starting node to the node, calculated according to the Euclidean distance formula. Suppose the starting node is (x1, y1), and the vertices under this node are (x i ,y i ),but min(L)={(L1),(L2),...(L n )}, min(L) is the optimal path to track the target passenger’s movement trajectory.

7. A passenger trajectory identification system, characterized in that: include: The first building module is used to build a monitoring camera network topology structure; The calculation module is used to divide the area according to the principle of human flow density based on the constructed surveillance camera network topology structure, and calculate the weight of each surveillance camera vertex based on the queuing theory model; The second construction module is used to perform vertex weighting operations on the constructed surveillance camera network topology structure according to the calculated vertex weights to construct a vertex weighted directed graph; The screening module is used to assign weights to vertices in the vertex weighted directed graph, and use an improved shortest path algorithm to comprehensively consider the weights between each vertex, historical path weights and path lengths, build a comprehensive evaluation function, screen the optimal video stream path, and determine the target passenger based on the optimal video stream path.

8. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium is used to store computer instructions, and when the computer instructions are executed by the processor, the passenger trajectory identification method according to any one of claims 1 to 6 is implemented.

9. A computer device, characterized in that: It includes a memory and a processor, the processor and the memory communicate with each other, the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the passenger trajectory identification method as described in any one of claims 1-6.

10. An electronic device, characterized in that: include: A processor, a memory and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory so that the electronic device executes instructions for implementing the passenger trajectory identification method as described in any one of claims 1-6.

Images