Monitoring video structured analysis method and system

Abstract

The invention discloses a monitoring video structured analysis method and system, and belongs to the technical field of video analysis, and the method comprises the steps: obtaining video data collected by all cameras, and recording a first camera as the camera of the level; determining the time when the target body enters the area from the video data corresponding to the camera at the level, recording the time as initial time, and performing video structured analysis on the video data corresponding to the camera at the level from the initial time to generate a video structured target description subfile; determining a camera corresponding to a road on which the target main body is about to go, and recording the camera as a lower-level camera; judging whether the video data corresponding to the lower-level camera contains a target main body or not; and if yes, performing video structured analysis on the video data corresponding to the lower-level camera, generating a video structured target description subfile, recording the lower-level camera as the camera of the level, and performing loop execution. The monitoring video structured analysis method and system are simple in operation and high in analysis efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of video analysis, and in particular to a method and system for structured analysis of surveillance videos. Background Art

[0002] As a crucial source of visual perception for the Internet of Things (IoT), video data is playing an increasingly crucial role in public safety. Video structured analysis, a deep application of unstructured video data, transforms it into perceptible, describable, and intelligent data, with a wide range of applications. For public safety, video structured analysis permeates nearly every aspect of public safety. Extracting relevant data through video structured analysis of video footage captured by surveillance cameras in public places is a common approach.

[0003] Currently, analysts typically need extensive experience and expertise, as well as the ability to operate video analysis software proficiently to accurately interpret and analyze complex surveillance videos. This requires filtering out information related to the target person from massive amounts of video data, eliminating interference factors, and ensuring the accuracy of the generated surveillance routes. This existing video analysis method is time-consuming and inefficient. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a surveillance video structured analysis method and system with simple operation and high analysis efficiency.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] In one aspect, the present invention provides a method for structured analysis of surveillance videos for monitoring an area, wherein a first camera is provided at the entrance of the area, and second to Nth cameras are provided on roads within the area, where N is an integer greater than 2. The method comprises:

[0007] Step S101: Obtain video data collected by all cameras, and record the first camera as the current-level camera;

[0008] Step S102: Determine the time when the target subject enters the area from the video data corresponding to the current camera, record it as the initial time, and perform video structured analysis on the video data corresponding to the current camera from the initial time to generate a video structured target description subfile;

[0009] Step S103: Determine the camera corresponding to the road that the target subject is about to travel, and record it as a subordinate camera;

[0010] Step S104: determining whether the video data corresponding to the lower-level camera contains the target subject;

[0011] Step S105: If included, perform video structured analysis on the video data corresponding to the subordinate camera, generate a video structured target description subfile, record the subordinate camera as the current level camera, and go to step S103.

[0012] In another aspect, the present invention provides a surveillance video structured analysis system, comprising:

[0013] An acquisition module, configured to acquire video data collected by all cameras and record the first camera as a current-level camera;

[0014] A first analysis module is configured to determine the time when the target subject enters the area from the video data corresponding to the current-level camera, record the time as the initial time, and perform video structured analysis on the video data corresponding to the current-level camera from the initial time to generate a video structured target description subfile;

[0015] A determination module is used to determine the camera corresponding to the road that the target subject is about to travel, which is recorded as a lower-level camera;

[0016] A judgment module, configured to judge whether the video data corresponding to the subordinate camera contains the target subject;

[0017] The second analysis module is used to perform video structured analysis on the video data corresponding to the subordinate camera, generate a video structured target description sub-file, and record the subordinate camera as the current level camera, and transfer it to the determination module.

[0018] On the other hand, the present invention provides an electronic device, comprising: a housing, a processor, a memory, a circuit board and a power supply circuit, wherein the circuit board is placed inside the space enclosed by the housing, and the processor and the memory are arranged on the circuit board; the power supply circuit is used to supply power to various circuits or devices of the above-mentioned electronic device; the memory is used to store executable program code; the processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the above-mentioned method.

[0019] In another aspect, the present invention provides a computer-readable storage medium storing one or more programs, wherein the one or more programs can be executed by one or more processors to implement the above method.

[0020] The present invention has the following beneficial effects:

[0021] The surveillance video structured analysis method and system of the present invention first records the first camera at the entrance of the area as the current-level camera, and then determines the camera corresponding to the road that the target subject is about to go to and records it as the lower-level camera. By promoting the lower-level camera that detects the target subject to the new current-level camera until the target disappears or leaves the entrance, continuous tracking of the target is ensured, and tracking interruptions caused by the target leaving the current camera's field of view are reduced; the tracking route of the camera is flexibly adjusted, that is, if the target is lost, the surrounding cameras are used for range search, so that the present invention can adapt to changes in the target movement path and enhance the flexibility of the system; only the target subject needs to be determined to obtain a video structured target description file about the target subject, which is simple to operate and has high analysis efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 Schematic diagram of the flow of the surveillance video structured analysis method of the present invention;

[0023] Figure 2 Schematic diagram of the principle of the surveillance video structured analysis method of the present invention;

[0024] Figure 3 This is a schematic diagram of the principle of determining the road that the target subject is about to travel in the present invention;

[0025] Figure 4 Schematic diagram of the structure of the surveillance video structured analysis system of the present invention;

[0026] Figure 5 Schematic diagram of the structure of the electronic device of the present invention. DETAILED DESCRIPTION

[0027] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0028] In the description of the present invention, it should be understood that the terms "up", "down", "front", "back", "left", "right", etc., indicating directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on the present invention.

[0029] On the one hand, the present invention provides a method for structured analysis of surveillance video, which is used to monitor a certain area, wherein a first camera is provided at the entrance of the area, and second to Nth cameras are provided on the roads within the area, where N is an integer greater than 2, such as Figure 1-2 As shown, the method includes:

[0030] Step S101: Obtain video data collected by all cameras, and record the first camera as the current-level camera;

[0031] In this step, all camera positions can be mapped to the map, with the current-level camera as the starting point for the target subject to appear, and the target subject's movement path in the area can be tracked. Since the first camera is located at the entrance of the area, the first camera can be used as the current-level camera to quickly check whether the target subject enters or leaves the area, thereby improving analysis efficiency.

[0032] As an optional embodiment, the area is a residential area, an amusement park, a zoo, or a factory. In a residential area or a factory, the method can be used to monitor the movements of strangers and protect the lives and property of the owners; in an amusement park or a zoo, the method can be used to find lost minors.

[0033] Step S102: Determine the time when the target subject enters the area from the video data corresponding to the current camera, record it as the initial time, and perform video structured analysis on the video data corresponding to the current camera from the initial time to generate a video structured target description subfile;

[0034] In this step, video structured analysis is performed on the video data corresponding to the current camera level, starting from the initial time. This can reduce the amount of data analysis and improve analysis efficiency. The target subject can be a person or a vehicle. Video structured analysis can be performed using conventional techniques in the field and will not be detailed here.

[0035] Step S103: Determine the camera corresponding to the road that the target subject is about to travel, and record it as a subordinate camera;

[0036] As an optional embodiment, the step S103 further includes: determining cameras corresponding to all roads connected to the road where the target subject is currently located, traversing these cameras, and recording them one by one as subordinate cameras.

[0037] In this step, the current-level camera is known, and the target subject needs to determine the lower-level camera to which the target subject is heading. In specific implementation, a graph structure can be used to represent the road network (nodes are road segments, edges are connection relationships), or an adjacency matrix, adjacency table, or graph database (such as Neo4j) can be used to manage camera relationships to quickly obtain the lower-level camera. It is understandable that among the multiple lower-level cameras thus determined, usually only one is correct. Therefore, this step traverses these cameras and, with the help of the subsequent step S104, determines whether the video data corresponding to the lower-level camera contains the target subject, thereby finding the correct lower-level camera (i.e., the lower-level camera to which the target subject is heading).

[0038] As another optional embodiment, step S103 includes:

[0039] Step S1031: determining the movement trajectory of the target subject according to the latest generated video structured target description subfile;

[0040] In this step, the video's structured object descriptor file is parsed to extract the target's location data. This location data includes the timestamp and the target's coordinates (x, y) at each time point. This extracted location data is then used to plot the target's movement path in two-dimensional space. This visualization can be achieved using tools such as Python combined with Matplotlib and OpenCV.

[0041] Step S1032: fitting the movement trajectory route according to the movement trajectory to obtain the movement trajectory extension line;

[0042] In this step, based on the known trajectory data, an algorithm (such as the least squares method, a spline fitting algorithm, or a Bezier curve, etc.) can be used to fit the movement trajectory extension line.

[0043] Step S1033: Determine the road that the target subject is about to travel based on the extended line of the movement trajectory.

[0044] The above steps S1031 to S1033 can be as follows Figure 3 As shown, Figure 3 In this example, the current camera is I, AB is the target's trajectory, OD, OE, and OF are the centerlines of the three roads connected to the target's current location, and BC is the extended line of the fitted trajectory. After determination, it is determined that the target's upcoming road is OE. In step 103, camera II is the determined lower-level camera. This method does not require traversing potential lower-level cameras, making lower-level camera determination highly efficient.

[0045] Preferably, the step S1033 includes:

[0046] Step S10331: for all roads connected to the road where the target subject is currently located, calculate the Manhattan distance between the center lines of all roads and the extended line of the moving trajectory;

[0047] In this step, the coordinates of the center line of the road corresponding to each camera can be pre-stored. The calculation formula of Manhattan distance is as follows:

[0048] d=|x1-x2|+|y1-y2|

[0049] Among them, x1 and y1 can be the coordinate points of the extension line of the movement trajectory, and x2 and y2 can be the coordinates of the center line of the road.

[0050] In specific implementation, during the calculation process, only the Manhattan distance between the center line of the corresponding road and the extended line of the moving trajectory can be calculated. To improve the accuracy of the calculation, several points (at least 3) can be selected on the center line of the corresponding road and the extended line of the moving trajectory, and their Manhattan distances can be calculated.

[0051] Step S10332: Determine the road corresponding to the minimum Manhattan distance as the road that the target subject is about to travel.

[0052] In this way, the road that the target subject is about to travel can be determined through steps S10331-step S10332, and the camera corresponding to the road that the target subject is about to travel can be determined through steps S1031-step S1033.

[0053] Step S104: determining whether the video data corresponding to the lower-level camera contains the target subject;

[0054] In specific implementation, a target detection model (such as YOLOv8, Faster R-CNN) can be used to detect whether the target subject exists.

[0055] Step S105: If included, perform video structured analysis on the video data corresponding to the subordinate camera, generate a video structured target description subfile, record the subordinate camera as the current level camera, and go to step S103.

[0056] In this step, video structured analysis is performed only on video data containing the target subject, which can speed up the video structured analysis and improve the efficiency of the video structured analysis.

[0057] In addition, if the subordinate camera is the first camera, it means that the target subject is about to leave the area. At this time, the complete path tracking of the target subject in the area has been achieved, and the entire process ends. Based on all video structured target description sub-files, the final video structured target description file is generated. The video structured target description file includes: target trajectory information, target attribute information and event information.

[0058] As an optional embodiment, the step S105 is further as follows:

[0059] If included, the time when the target subject leaves the current-level camera is taken as the starting time, and video structured analysis is performed on the video data corresponding to the lower-level camera from the starting time to generate a video structured target description sub-file, and the lower-level camera is recorded as the current-level camera, and go to step S103.

[0060] In this embodiment, performing video structured analysis on the video data corresponding to the lower-level camera from the initial time can further reduce the amount of data analysis and improve analysis efficiency.

[0061] As another optional embodiment, the step S104 includes:

[0062] Step S105': If not, searching for all cameras within a preset range excluding the lower-level camera with the current-level camera as the center, and recording them as peripheral cameras; determining whether the video data corresponding to the peripheral cameras contains the target subject;

[0063] In this step, it is known that the target subject last appeared in the field of view of the camera at this level and has a departure time point (i.e., the start time). With the camera at this level as the center, all cameras except the lower-level camera within a preset range (which can be flexibly set according to needs, such as 200m, 300m, 500m, etc.) are searched to find the target subject again.

[0064] Step S106': If the video data corresponding to the surrounding cameras contains the target subject, the time when the target subject leaves the current-level camera is taken as the starting time, and from this starting time, the video data corresponding to the surrounding cameras containing the target subject are subjected to video structured analysis to generate a video structured target description subfile, and the surrounding cameras containing the target subject are recorded as current-level cameras, and the process goes to step S103.

[0065] In this step, when the lower-level camera loses the target subject, the surrounding cameras are used to search for the target subject to deal with situations where the target subject suddenly turns to other directions or maliciously avoids the camera, etc., which is flexible and easy to use.

[0066] Preferably, the step S105' includes:

[0067] Step S106": If the video data corresponding to the surrounding cameras do not contain the target subject, a video structured target description file is generated based on all video structured target description subfiles, and the video structured target description file includes: target trajectory information, target attribute information and event information.

[0068] In this step, all video structured description sub-files related to the target subject may be integrated according to the generation time to generate a final video structured target description (total) file.

[0069] In specific implementation, redundant and erroneous data can be removed first; then, missing parts of the trajectory are interpolated and completed; finally, inconsistent attributes are fused using confidence-weighted methods. To facilitate analysis, the generated video structured target description file is stored in a database (such as Elasticsearch, MongoDB, or a time series database). The generated video structured target description file supports rapid retrieval based on time, location, target ID, event type, and other criteria. Furthermore, the video structured target description file can be exported to various formats such as PDF, CSV, and XML for easy reading or export.

[0070] In the embodiment of the present invention, preferably, the target subject is a person, and in the method, the method for determining whether the video data contains the target subject includes the steps of:

[0071] A1: Divide the target subject's feature information into facial information, height information, and clothing color, thereby generating height information comparison tasks, clothing color comparison tasks, and facial information comparison tasks;

[0072] A2: Execute the height information comparison task, clothing color comparison task, and face information comparison task in sequence, using the comparison result obtained from the previous comparison task as the comparison range for the next comparison task, until all comparison tasks are completed.

[0073] In specific implementation, as shown in Table 1, the first step is to compare height information, which can be measured by visual estimation or lidar. The second step is to compare clothing colors within the height comparison results, which can specifically include top color, pants color, etc., and can be further screened by the presence or absence of a backpack and the backpack color. The third step is to compare facial information within the clothing color comparison results, which can specifically include facial feature vectors. To accelerate the comparison, gender and age estimation can also be performed.

[0074] In this way, after the previous single feature comparison task (i.e., one of the height information comparison task, clothing color comparison task, and face information comparison task) is completed, the next single feature comparison task is executed based on the result returned by the previous single feature comparison task. This narrows the scope of the comparison of the next single feature comparison task, and the feature information volume of the three single feature comparison tasks is arranged from small to large, which can greatly improve efficiency while ensuring accuracy.

[0075] Table 1

[0076]

[0077] In summary, the surveillance video structured analysis method of the present invention is used to monitor a certain area, where a first camera is provided at the entrance of the area, and second to N-th cameras are provided on the roads within the area, where N is an integer greater than 2. The method includes: obtaining video data collected by all cameras, and recording the first camera as the current-level camera; determining the time when the target subject enters the area from the video data corresponding to the current-level camera, recording it as the initial time, and performing video structured analysis on the video data corresponding to the current-level camera from the initial time to generate a video structured target description subfile; determining the camera corresponding to the road to which the target subject is about to go, recording it as a lower-level camera; judging whether the video data corresponding to the lower-level camera contains the target subject; if so, performing video structured analysis on the video data corresponding to the lower-level camera, generating a video structured target description subfile, and recording the lower-level camera as the current-level camera, and executing the operation in a loop.

[0078] The method of the present invention first records the first camera at the entrance of the area as the current-level camera, and then determines the camera corresponding to the road that the target subject is about to go to and records it as the lower-level camera. By promoting the lower-level camera that detects the target subject to the new current-level camera until the target disappears or leaves the entrance, continuous tracking of the target is ensured, and tracking interruptions caused by the target leaving the current camera's field of view are reduced; the tracking route of the camera is flexibly adjusted, that is, if the target is lost, the surrounding cameras are used for range search, so that the present invention can adapt to changes in the target movement path and enhance the flexibility of the system; only the target subject needs to be determined to obtain a video structured target description file about the target subject, which is simple to operate and has high analysis efficiency.

[0079] The present invention is particularly suitable for tracking the path of a selected target subject after entering a certain area (until lost or leaving the area); it is conceivable that the path can also be tracked in reverse chronological order with reference to the principle of the present invention, starting from the target subject leaving the area.

[0080] This method not only enables precise and continuous tracking of target entities, but also enhances the overall system's intelligence and practical application value through structured data accumulation, multi-dimensional feature comparison, and flexible tracking methods. It can provide businesses and individuals with powerful tools to manage and analyze complex video surveillance data, thereby better serving various business needs and social security requirements.

[0081] On the other hand, the present invention provides a surveillance video structured analysis system, such as Figure 4 As shown, the system includes:

[0082] An acquisition module 11 is configured to acquire video data collected by all cameras and record the first camera as a current-level camera;

[0083] A first analysis module 12 is configured to determine the time when the target subject enters the area from the video data corresponding to the current camera, record the time as the initial time, and perform video structured analysis on the video data corresponding to the current camera from the initial time to generate a video structured target description subfile;

[0084] A determination module 13 is used to determine the camera corresponding to the road that the target subject is about to travel, which is recorded as a lower-level camera;

[0085] A judgment module 14 is used to judge whether the video data corresponding to the lower-level camera contains the target subject;

[0086] The second analysis module 15 is used to perform video structured analysis on the video data corresponding to the subordinate camera, generate a video structured target description sub-file, and record the subordinate camera as the current level camera, and transfer to the determination module 13.

[0087] The device of this embodiment can be used to perform Figure 1 The technical solution of the method embodiment shown has similar implementation principles and technical effects, which will not be repeated here.

[0088] Preferably, the determining module 13 is further configured to determine cameras corresponding to all roads connected to the road where the target subject is currently located, traverse these cameras, and record them one by one as subordinate cameras.

[0089] Preferably, the determining module 13 includes:

[0090] The judgment submodule 131 is used to determine the movement trajectory of the target subject according to the latest generated video structured target description subfile;

[0091] A fitting submodule 132 is configured to fit the movement trajectory according to the movement trajectory to obtain an extension line of the movement trajectory;

[0092] The determination submodule 133 is configured to determine the road that the target subject is about to travel to based on the extended line of the movement trajectory.

[0093] Preferably, the determining submodule 133 includes:

[0094] A calculation unit 1331 is configured to calculate, for all roads connected to the road currently located by the target subject, the Manhattan distance between the center lines of all the roads and the extended line of the moving trajectory;

[0095] The determining unit 1332 is configured to determine the road corresponding to the minimum Manhattan distance as the road that the target subject is about to travel.

[0096] Preferably, the second analysis module 15 is further used to, if included, take the time when the target subject leaves the current-level camera as the starting time, perform video structured analysis on the video data corresponding to the lower-level camera from the starting time, generate a video structured target description sub-file, and record the lower-level camera as the current-level camera, and go to the determination module 13.

[0097] Preferably, the system further comprises:

[0098] The third analysis module 15' is configured to, if the target subject is not included, search for all cameras within a preset range excluding the current camera and the lower-level cameras, with the current camera as the center, and record them as peripheral cameras; and determine whether the video data corresponding to the peripheral cameras contains the target subject;

[0099] The fourth analysis module 16' is used to perform video structured analysis on the video data corresponding to the surrounding cameras containing the target subject if the video data corresponding to the surrounding cameras contains the target subject, taking the time when the target subject leaves the current-level camera as the starting time, generating a video structured target description sub-file, and recording the surrounding cameras containing the target subject as current-level cameras, and transferring the file to the determination module 13.

[0100] Preferably, the system further comprises:

[0101] The fifth analysis module 16 is used to generate a video structured target description file based on all video structured target description subfiles if the video data corresponding to the surrounding cameras do not contain the target subject. The video structured target description file includes: target trajectory information, target attribute information and event information.

[0102] Preferably, the target subject is a person, and in the system, the method for determining whether the video data contains the target subject includes:

[0103] The target subject feature information is divided into facial information, height information and clothing color, thereby generating height information comparison tasks, clothing color comparison tasks and facial information comparison tasks;

[0104] The height information comparison task, clothing color comparison task and face information comparison task are executed in sequence, and the comparison result obtained from the previous comparison task is used as the comparison range of the next comparison task until all comparison tasks are completed.

[0105] Preferably, the area is a residential area, an amusement park, a zoo or a factory.

[0106] The present invention also provides an electronic device, Figure 5This is a schematic diagram of the structure of the electronic device of the present invention, which can realize the present invention Figure 1 The process shown, such as Figure 5 As shown, the above-mentioned electronic device may include: a shell 41, a processor 42, a memory 43, a circuit board 44 and a power supply circuit 45, wherein the circuit board 44 is placed inside the space enclosed by the shell 41, and the processor 42 and the memory 43 are arranged on the circuit board 44; the power supply circuit 45 is used to supply power to various circuits or devices of the above-mentioned electronic device; the memory 43 is used to store executable program code; the processor 42 runs the program corresponding to the executable program code by reading the executable program code stored in the memory 43, so as to execute the method described in any of the above-mentioned method embodiments.

[0107] For details on the specific execution process of the above steps by the processor 42 and the steps further executed by the processor 42 by running the executable program code, please refer to the present invention. Figure 1 The description of the illustrated embodiment will not be repeated here.

[0108] This electronic device exists in many forms, including but not limited to:

[0109] (1) Mobile communication devices: These devices are characterized by their mobile communication capabilities and are primarily designed to provide voice and data communications. These terminals include smartphones (e.g., iPhones), multimedia phones, feature phones, and low-end phones.

[0110] (2) Ultra-mobile personal computer devices: These devices fall under the category of personal computers, have computing and processing capabilities, and generally also have mobile Internet access. These terminals include PDAs, MIDs, and UMPCs, such as the iPad.

[0111] (3) Portable entertainment devices: These devices can display and play multimedia content. These devices include audio and video players (such as iPods), handheld game consoles, e-books, smart toys, and portable car navigation devices.

[0112] (4) Server: A device that provides computing services. The server consists of a processor, hard disk, memory, system bus, etc. The server is similar to a general computer architecture, but because it needs to provide highly reliable services, it has higher requirements in terms of processing power, stability, reliability, security, scalability, and manageability.

[0113] (5) Other electronic devices with data interaction functions.

[0114] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method steps described in any of the above method embodiments are implemented.

[0115] The present invention also provides an application program, which is executed to implement the method provided by any method embodiment of the present invention.

[0116] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0117] Each embodiment in this specification is described in a related manner. The same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple. For the relevant parts, refer to the partial description of the method embodiment. For the convenience of description, the above device is described by dividing it into various units / modules according to their functions. Of course, when implementing the present invention, the functions of each unit / module can be implemented in the same or multiple software and / or hardware.

[0118] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware through a computer program. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0119] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for structured analysis of surveillance video, for monitoring an area, wherein a first camera is provided at the entrance of the area, and second to Nth cameras are provided on roads within the area, where N is an integer greater than 2, characterized in that: The method comprises: Step S101: Obtain video data collected by all cameras, and record the first camera as the current-level camera; Step S102: Determine the time when the target subject enters the area from the video data corresponding to the current camera, record it as the initial time, and perform video structured analysis on the video data corresponding to the current camera from the initial time to generate a video structured target description subfile; Step S103: Determine the camera corresponding to the road that the target subject is about to travel, and record it as a subordinate camera; Step S104: determining whether the video data corresponding to the lower-level camera contains the target subject; Step S105: If included, perform video structured analysis on the video data corresponding to the subordinate camera, generate a video structured target description subfile, record the subordinate camera as the current level camera, and go to step S103.

2. The surveillance video structured analysis method according to claim 1, characterized in that: The step S103 further includes: determining cameras corresponding to all roads connected to the road where the target subject is currently located, traversing these cameras, and recording them one by one as subordinate cameras.

3. The method for structured analysis of surveillance video according to claim 1, characterized in that: The step S103 includes: Step S1031: determining the movement trajectory of the target subject according to the latest generated video structured target description subfile; Step S1032: fitting the movement trajectory route according to the movement trajectory to obtain the movement trajectory extension line; Step S1033: Determine the road that the target subject is about to travel based on the extended line of the movement trajectory.

4. The method for structured analysis of surveillance video according to claim 3, characterized in that: The step S1033 includes: Step S10331: for all roads connected to the road where the target subject is currently located, calculate the Manhattan distance between the center lines of all roads and the extended line of the moving trajectory; Step S10332: Determine the road corresponding to the minimum Manhattan distance as the road that the target subject is about to travel.

5. The method for structured analysis of surveillance video according to claim 1, characterized in that: The step S105 is further as follows: If included, the time when the target subject leaves the current-level camera is taken as the starting time, and video structured analysis is performed on the video data corresponding to the lower-level camera from the starting time to generate a video structured target description sub-file, and the lower-level camera is recorded as the current-level camera, and go to step S103.

6. The method for structured analysis of surveillance video according to claim 1, characterized in that: The step S104 includes: Step S105': If not, searching for all cameras within a preset range except the lower-level cameras with the current-level camera as the center, and recording them as peripheral cameras; determining whether the video data corresponding to the peripheral cameras contains the target subject; Step S106': If the video data corresponding to the surrounding cameras contains the target subject, the time when the target subject leaves the current-level camera is taken as the starting time, and from this starting time, the video data corresponding to the surrounding cameras containing the target subject are subjected to video structured analysis to generate a video structured target description subfile, and the surrounding cameras containing the target subject are recorded as current-level cameras, and the process goes to step S103.

7. The method for structured analysis of surveillance video according to claim 6, characterized in that: The step S105' then includes: Step S106": If the video data corresponding to the surrounding cameras do not contain the target subject, a video structured target description file is generated based on all video structured target description subfiles, and the video structured target description file includes: target trajectory information, target attribute information and event information.

8. The method for structured analysis of surveillance video according to claim 1, wherein: The target subject is a person. In the method, the method for determining whether the video data contains the target subject includes: The target subject feature information is divided into facial information, height information and clothing color, thereby generating height information comparison tasks, clothing color comparison tasks and facial information comparison tasks; The height information comparison task, clothing color comparison task and face information comparison task are executed in sequence, and the comparison result obtained from the previous comparison task is used as the comparison range of the next comparison task until all comparison tasks are completed.

9. The method for structured analysis of surveillance video according to claim 1, wherein: The area is a residential area, an amusement park, a zoo or a factory.

10. A surveillance video structured analysis system, characterized in that: The system comprises: An acquisition module, configured to acquire video data collected by all cameras and record the first camera as a current-level camera; A first analysis module is configured to determine the time when the target subject enters the area from the video data corresponding to the current-level camera, record the time as the initial time, and perform video structured analysis on the video data corresponding to the current-level camera from the initial time to generate a video structured target description subfile; A determination module is used to determine the camera corresponding to the road that the target subject is about to go, which is recorded as a lower-level camera; A judgment module, configured to judge whether the video data corresponding to the subordinate camera contains the target subject; The second analysis module is used to perform video structured analysis on the video data corresponding to the subordinate camera, generate a video structured target description sub-file, and record the subordinate camera as the current level camera, and transfer it to the determination module.

Images