Vehicle cross-boundary tracking method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of intelligent traffic, in particular to a vehicle cross-boundary tracking method and device, electronic equipment and a storage medium, the method is applied to a vehicle tracking system, and the vehicle tracking system comprises a management platform and an algorithm platform which are connected in sequence; the method comprises the following steps: through a configuration rule line of a management platform, establishing a logic pairing relationship between an entry rule line and an exit rule line, so as to generate an algorithm task containing a camera ID, a rule line coordinate and a region pairing relationship; and the algorithm platform receives the algorithm task, performs vehicle cross-border tracking based on configuration data of the algorithm task, and feeds back a tracking result to the management platform. The monitoring rule is configured based on the management platform, the monitoring platform is controlled to perform vehicle cross-border tracking based on the monitoring rule under the action of the task platform, vehicle cross-border tracking in a complex scene can be adapted, the track integrity is improved, and the error rate is reduced.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a vehicle cross-border tracking method, device, electronic equipment, and storage medium. Background Art

[0002] With the development of economy and urbanization, the number of trade vehicles in bonded areas has increased sharply, and traditional traffic management methods are difficult to meet the needs of refined supervision.

[0003] Thanks to advances in deep learning and computer vision technology, the accuracy and efficiency of vehicle detection and tracking have significantly improved. Simultaneously, declining hardware costs and increased computing power have made large-scale applications a reality. The availability of massive amounts of traffic video data and public datasets provides a solid foundation for algorithm training and verification; advances in data annotation technology further enhance training efficiency and effectiveness. Existing vehicle tracking methods primarily rely on single-camera target detection and tracking technology, recording vehicle motion trajectories through frame-to-frame matching and feature extraction.

[0004] However, existing technologies cannot adapt to vehicle tracking in complex scenes, and the tracking error rate is high. Summary of the Invention

[0005] In view of this, an object of the present invention is to provide a vehicle cross-border tracking method, device, electronic device, and storage medium.

[0006] In a first aspect, an embodiment of the present invention provides a vehicle cross-border tracking method, which is applied to a vehicle tracking system. The vehicle tracking system includes a management platform and an algorithm platform connected in sequence. The method includes:

[0007] By configuring rule lines on the management platform, and establishing a logical pairing relationship between entry rule lines and exit rule lines, an algorithm task is generated that includes camera ID, rule line coordinates, and area pairing relationship.

[0008] The algorithm platform receives the algorithm tasks, performs cross-border vehicle tracking based on the configuration data of the algorithm tasks, and feeds back the tracking results to the management platform.

[0009] In conjunction with the first aspect, the management platform provides a graphical display interface; the steps of configuring monitoring rules through the management platform include:

[0010] In response to the addition operation for monitoring, the video stream corresponding to the added monitoring device code is displayed on the bottom canvas interactive interface;

[0011] In response to the monitoring rule configuration operation, the upper transparent canvas interaction interface is displayed in a floating manner on the bottom canvas interaction interface;

[0012] In response to click operations on multiple points in the upper transparent canvas interactive interface, each point is treated as a regular point and connected in sequence according to the click sequence to form a closed polygon;

[0013] The closed polygon is used as the monitoring area, the vehicles in the monitoring area are used as the vehicles to be monitored, and the codes of the monitoring devices corresponding to the monitoring area are used as the monitoring device codes to obtain the monitoring rules.

[0014] In conjunction with the first aspect, the steps for configuring monitoring rules through the management platform include:

[0015] In response to the triggering operation of the first designated key, all rule points and the currently constructed monitoring area in the upper transparent canvas interaction interface are deleted.

[0016] In conjunction with the first aspect, after configuring monitoring rules through the management platform, the following steps are also included:

[0017] In response to the drag operation on the target rule point, the current position, monitoring area, and monitoring device code of the target rule point are updated to obtain an updated monitoring rule.

[0018] In conjunction with the first aspect, the step of establishing a logical pairing relationship between an entry rule line and an exit rule line includes:

[0019] In response to the area association instruction, the entry rule line and the exit rule line are bound into a pairing group;

[0020] Set a unique region identifier for the pairing group and associate it with the camera ID.

[0021] In conjunction with the first aspect, the steps for cross-border vehicle tracking based on the configuration data of the algorithm task include:

[0022] A deep learning detection model is used to detect vehicles in real time in video frames corresponding to the configured rule lines, generating vehicle bounding box information with timestamps.

[0023] Based on the vehicle bounding box information, the tracking algorithm associates the vehicle targets in consecutive video frames to generate the vehicle motion trajectory within a single camera;

[0024] Match vehicle trajectories with depth feature similarity above a threshold and spatiotemporal continuity from different cameras and fit the global motion trajectory through coordinate transformation;

[0025] When the global trajectory triggers the paired entry rule line and exit rule line in sequence, it is marked as a cross-border event.

[0026] In conjunction with the first aspect, when the global trajectory triggers the paired entry rule line and exit rule line in sequence, after marking it as a cross-border event, the following steps are also included:

[0027] Intercept the video clip from t seconds before the entry rule line is triggered to t seconds after the exit rule line is triggered;

[0028] Feedback cross-border event data, global trajectories, and video clips to the management end for storage.

[0029] In a second aspect, embodiments of the present application further provide a vehicle cross-border tracking device, which is applied to a vehicle tracking system. The vehicle tracking system includes a management platform and an algorithm platform connected in sequence. The device includes:

[0030] The configuration module is used to configure the rule lines through the management platform and establish a logical pairing relationship between the entry rule line and the exit rule line to generate an algorithm task containing the camera ID, rule line coordinates and area pairing relationship;

[0031] The tracking module is used for the algorithm platform to receive algorithm tasks, conduct cross-border vehicle tracking based on the configuration data of the algorithm tasks, and feed back the tracking results to the management platform.

[0032] In a third aspect, the present application provides an electronic device, which includes a memory and a processor. The memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the above method.

[0033] In a fourth aspect, the present application provides a readable storage medium, in which computer program instructions are stored. When the computer program instructions are read and executed by a processor, the above-mentioned method is executed.

[0034] The embodiments of the present invention bring the following beneficial effects: the vehicle cross-border tracking method, device, electronic device, and storage medium provided by the present application are applied to a vehicle tracking system, and the vehicle tracking system includes a management platform and an algorithm platform connected in sequence; the method includes: configuring rule lines through the management platform, and establishing a logical pairing relationship between entry rule lines and exit rule lines to generate an algorithm task including a camera ID, rule line coordinates, and area pairing relationship; the algorithm platform receives the algorithm task, and performs cross-border vehicle tracking based on the configuration data of the algorithm task, and feeds back the tracking results to the management platform.

[0035] This application configures monitoring rules based on the management platform, and controls the monitoring platform to track vehicles across borders based on the monitoring rules under the action of the task platform. It can adapt to cross-border vehicle tracking in complex scenarios, improve trajectory integrity and reduce error rate.

[0036] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or understood by practicing the present invention. The purposes and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0037] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without paying any creative work.

[0039] Figure 1 A schematic flow chart of a vehicle cross-border tracking method provided by an embodiment of the present invention;

[0040] Figure 2 An example diagram of a graphical display interface provided by the management platform according to an embodiment of the present invention;

[0041] Figure 3 This is an example diagram of the monitoring point management interface provided by an embodiment of the present invention;

[0042] Figure 4 This is an example diagram of the monitoring point management interface provided by an embodiment of the present invention;

[0043] Figure 5 A schematic diagram of a vehicle cross-border tracking device provided by an embodiment of the present invention;

[0044] Figure 6 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention.

[0045] Reference numerals:

[0046] 10-Configuration module, 20-Tracking module;

[0047] 130 - processor, 131 - memory, 132 - bus, 133 - communication interface. DETAILED DESCRIPTION

[0048] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work shall fall within the scope of protection of the present invention.

[0049] To facilitate understanding of this embodiment, the application scenarios and design concepts of the embodiment of this application are briefly introduced below.

[0050] In complex traffic scenarios, existing vehicle tracking methods have many limitations: first, factors such as occlusion, lighting changes, and dynamic background interference can easily lead to target loss or misidentification; second, the lack of a cross-camera collaboration mechanism makes it difficult to achieve full-domain perception and ID consistency maintenance, resulting in frequent "ID drift" phenomena; in addition, the existing system's judgment of rule line crossing events is mostly based on static settings and spatial positions, and does not fully combine time series and trajectory prediction models, resulting in imprecise event judgment logic, high false alarm and missed alarm rates, and insufficient overall tracking accuracy and stability.

[0051] Based on this, embodiments of the present application provide a vehicle cross-border tracking method, device, electronic device, and storage medium.

[0052] Example 1

[0053] The present application provides a vehicle cross-border tracking method, which is applied to a vehicle tracking system. The vehicle tracking system includes a management platform and an algorithm platform connected in sequence. Figure 1 As shown, the method includes:

[0054] S110, configuring rule lines through the management platform, and establishing a logical pairing relationship between entry rule lines and exit rule lines to generate an algorithm task including camera ID, rule line coordinates and area pairing relationship.

[0055] S120, the algorithm platform receives the algorithm task, performs cross-border vehicle tracking based on the configuration data of the algorithm task, and feeds back the tracking results to the management platform.

[0056] The vehicle cross-border tracking method provided in this application is based on the rule lines configured by the user, and after establishing a tracking area divided by the entry and exit rule lines, the algorithm task is sent to the algorithm platform to obtain the video stream and process the data of the tracking area to track the vehicle.

[0057] In conjunction with the first aspect, the management platform provides a graphical display interface. The step of configuring monitoring rules through the management platform in step S110 includes:

[0058] S111 , in response to the adding operation for monitoring, displaying the video stream corresponding to the added monitoring device encoding on the bottom canvas interactive interface.

[0059] S112 , in response to the monitoring rule configuration operation, the upper transparent canvas interaction interface is displayed floating on the bottom canvas interaction interface.

[0060] S113 , in response to click operations on multiple points in the upper transparent canvas interactive interface, each point is regarded as a regular point and connected in sequence according to the click sequence to form a closed polygon.

[0061] S114, taking the closed polygon as a monitoring area, taking a vehicle in the monitoring area as a to-be-monitored vehicle, taking an encoding of a monitoring device corresponding to the monitoring area as a monitoring device encoding, and obtaining a monitoring rule.

[0062] Please refer to Figure 2 , an exemplary graphical display interface is provided, a menu bar is provided on the graphical display interface, a vehicle cross-border tracking module is provided in the menu bar, the vehicle cross-border tracking module is a parent node, a plurality of child nodes (such as a monitoring point management module, a monitoring area management, an algorithm task management module, a task monitoring association module, etc.) corresponding to the parent node, when a user selects or clicks the monitoring point management module, the graphical display interface is rendered, and a monitoring point management interface as shown in Figure 3 is generated. In the monitoring point management interface, a list containing a plurality of monitoring point information is provided, and a bottom layer canvas interactive interface on the right side of the list and an add module above the list (in combination with Figure 3 “add” shown). After one of the monitoring points in the list, the video stream collected by the shooting device corresponding to the monitoring point is loaded on the bottom layer canvas interactive interface on the right side.

[0063] A configuration module (such as “draw area” shown in Figure 3 ) is also provided below the bottom layer canvas interactive interface, when a user clicks the module to configure a monitoring rule, a top layer transparent canvas interactive interface is displayed on the bottom layer canvas interactive interface.

[0064] Subsequently, the user clicks on the top layer transparent canvas interactive interface in sequence, and sequentially connects the lines according to the click time sequence to form a closed polygon, so as to frame out the to-be-tracked vehicle, the user can freely draw on the transparent canvas, the system will recognize each click of the user as a rule point, and automatically connect the lines according to the click order to form a closed polygon contour, for example Figure 3 the closed pentagon in. The polygon is the set monitoring area, the system marks the vehicle appearing in the area as a to-be-monitored object, and binds the area with the corresponding camera device encoding to generate a complete monitoring rule.

[0065] Among them, after the user triggers (clicks or touches to select) the new module, the new monitoring point information is added to the list. Specifically, it includes: after the user triggers (clicks or touches to select) the new module, a new monitoring point configuration window pops up, and an input area is provided in the configuration window for entering or selecting the monitoring device code, device name, installation location and associated video stream address; receiving the monitoring point information entered by the user, and saving it to the monitoring point information list, and at the same time linking it with the underlying canvas interaction interface; after saving the monitoring point information, automatically loading the corresponding video stream to the underlying canvas interaction interface, so that the user can view the screen content of the monitoring point in real time; synchronizing the newly added monitoring point information to the algorithm platform, so that the algorithm platform can perform task allocation and vehicle tracking processing based on the latest monitoring point data. Through the above steps, the dynamic addition and instant visual display of monitoring points are realized, which improves the operability and real-time performance of the system.

[0066] In conjunction with the first aspect, the step of configuring monitoring rules through the management platform in step S110 further includes:

[0067] S115 , in response to the triggering operation of the first designated key, deleting all the rule points and the currently constructed monitoring area in the upper transparent canvas interaction interface.

[0068] In this embodiment, in order to improve the flexibility and fault tolerance of users when configuring monitoring areas, the system provides a mechanism for quickly clearing the currently drawn content. When the user is drawing rule points in the upper transparent canvas interactive interface, if a drawing error is found, the area selection is inappropriate, or the monitoring range needs to be readjusted, the user can click the "Clear" button on the graphic display interface or press the preset shortcut key (such as "Ctrl+Z" or "Esc" key), and the system will immediately clear all drawn rule points and the monitoring area outline composed of these rule points. This function supports users to roll back multiple operations, and can re-enter the drawing state after clearing to continue to complete the new monitoring area configuration. At the same time, the system retains a historical drawing record cache, allowing recovery to any drawing state before the user confirms the save, thereby improving user operation efficiency and experience.

[0069] In conjunction with the first aspect, after the step of configuring monitoring rules through the management platform in S110, the following steps may be further included:

[0070] S116 , in response to the drag operation on the target rule point, the current position, monitoring area, and monitoring device code of the target rule point are updated to obtain an updated monitoring rule.

[0071] In this embodiment, to enhance user flexibility and interactive experience in monitoring area configuration, the upper-layer transparent canvas interface dynamically adjusts drawn rule points. Specifically, after the user completes the initial closed polygon drawing, they can hover the cursor over a rule point (i.e., the target rule point) and, by pressing the left mouse button, drag the point to a new location. The system detects this drag operation in real time and updates the entire monitoring area shape based on the target rule point's new coordinates. Simultaneously, the system automatically and synchronously updates the monitoring device coding information associated with the monitoring area to ensure the accuracy and consistency of the monitoring rules. The updated monitoring rules include the latest set of rule point coordinates, the revised monitoring area range, and the corresponding device coding information. Furthermore, "Save" and "Cancel" options are provided below the lower-layer canvas interface for users to choose whether to retain the changes. If the user selects "Save," the updated monitoring rules will be submitted to the algorithm platform for use in subsequent cross-border vehicle tracking tasks; if the user selects "Cancel," the system rolls back to its original state. This feature effectively enhances users' ability to fine-tune their monitoring area and enhances the system's usability and user-friendliness.

[0072] In combination with the first aspect, step S110 of establishing a logical pairing relationship between an entry rule line and an exit rule line includes:

[0073] S117 , responding to the area association instruction, binding the entry rule line and the exit rule line into a pairing group.

[0074] S118 , setting a unique area identifier for the pairing group and associating it with the camera ID.

[0075] In this embodiment, the area association instruction can be initiated by the user on the graphical display interface by dragging, selecting a box, or clicking a confirmation button. In combination with the above example, the area association instruction can be initiated by clicking "End Drawing" at the bottom of the underlying canvas interaction interface.

[0076] After the system receives the instruction, Figure 3 A pop-up window is generated on the monitoring point management interface shown (combined with Figure 4 As shown in the figure, the user edits the area name and configures the attribute information (specifically, the entry rule line and exit rule line). The spatial position relationship between the entry rule line and the exit rule line is then verified to ensure that there is a reasonable logical relationship between the two (such as being located in the same monitoring screen and facing each other). Once the pairing is successful, the system will highlight the pairing group in the graphical display interface and save its metadata such as the area identifier, camera ID, rule line coordinate information, and pairing establishment time in the background database for subsequent query, modification, or deletion operations.

[0077] In addition, the system also provides editing or unbinding for the established pairing group. For example, the user can disconnect the logical association between a certain entry rule line and an exit rule line by selecting "unpair" in the right-click menu, and the system will synchronously update the mapping relationship between the area identifier and the camera ID, avoiding affecting the data integrity of other valid pairing groups. This mechanism improves the logical management capability of the system and helps to realize fine identification and tracking of vehicle cross-border behavior in complex monitoring scenarios.

[0078] It can be understood that the algorithm data configured needs to be referred to when generating the algorithm task, combined with the Figure 2-Figure 4 The algorithm task management module is shown in the figure. The user clicks the module to display the algorithm management interface (not shown in the figure) on the graphical display interface as shown in Figure 2 In the interface, target algorithm selection, algorithm parameter configuration, and algorithm addition and deletion can be performed, which will not be described here.

[0079] In combination with the first aspect, the step of tracking the vehicle cross-border based on the configuration data of the algorithm task in step S120 includes:

[0080] S121, using a deep learning detection model to detect vehicles in video frames corresponding to the configuration rule line in real time, generating time-stamped vehicle bounding box information.

[0081] After planning the monitoring area by the configuration rule line of the management platform, the vehicle information in the video corresponding to the monitoring area is acquired by the deep learning detection model. Common deep learning detection models include YOLO series models, SSD, and Faster R-CNN. These models identify vehicle targets in each frame of image by convolution, feature extraction, and classification judgment, and draw a bounding box for each detected vehicle to represent its position coordinates (x, y, width, height) in the picture. At the same time, the timestamp of the frame is recorded for subsequent tracking.

[0082] S122, based on the vehicle bounding box information, associating the vehicle targets in consecutive video frames by a tracking algorithm to generate the vehicle motion trajectory in a single camera.

[0083] After completing the target detection, in order to obtain the motion path of the vehicle in the video sequence, the system uses a multi-target tracking algorithm to cross-frame associate the detection results to construct the motion trajectory of the vehicle. Commonly used tracking algorithms include SORT (Simple Online and Realtime Tracking), DeepSOR or ByteTrack. By specifying a certain tracking algorithm, the boundary box information and the corresponding target ID of the vehicle in each frame are obtained, and these information are organized in time sequence into a track (Track), which contains the following information: target ID, timestamp list, coordinate track point set, appearance feature vector (such as color, vehicle model, etc.), so as to form the vehicle motion trajectory in the field of view of a single camera.

[0084] S123, matching vehicle trajectories with high similarity in different cameras and spatiotemporal continuity and fitting global motion trajectories through coordinate transformation.

[0085] In order to realize the cross-camera vehicle trajectory splicing and global tracking, feature extraction and comparison, spatiotemporal consistency verification, coordinate transformation and trajectory fusion are specifically implemented.

[0086] Among them, the feature extraction and comparison includes: using convolutional neural network (CNN) or Transformer architecture to extract the appearance features of the vehicle, such as color, vehicle model, texture, etc.; comparing the feature vectors of the vehicles in different cameras, calculating their similarity (such as cosine similarity), and finding the most likely matching item; introducing attention mechanism to improve the attention of key features and enhance the feature discrimination ability; using transfer learning strategy to speed up the training process and reduce the computational overhead.

[0087] The spatiotemporal consistency verification includes: combining the timestamps of the vehicles entering and leaving each camera, the driving direction, the estimated speed, etc., to verify whether the trajectory has spatiotemporal continuity; excluding the cases where the appearance is similar but actually not the same vehicle.

[0088] The coordinate transformation and trajectory fusion includes: using the relative position relationship between the cameras and the spatial mapping matrix to convert the trajectories of different perspectives to a unified global coordinate system; fitting all matched local trajectories to form a complete global vehicle motion trajectory,

[0089] S124, when the global trajectory triggers the paired entry rule line and exit rule line in turn, it is marked as a cross-border event.

[0090] In this embodiment, whether the vehicle on the global trajectory crosses the "entry rule line" and "exit rule line" defined by the user is detected. The specific process includes trajectory crossing judgment, cross-border event generation, and event display and storage.

[0091] Among them, the trajectory crossing judgment specifically includes:

[0092] Determine whether the vehicle's trajectory intersects with the entry rule line (entry behavior); then determine whether the trajectory intersects with the paired exit rule line (exit behavior); the system ensures the correct crossing order between the two to avoid misjudgment (such as only entering but not leaving).

[0093] Cross-border event generation specifically includes: if the vehicle trajectory completely crosses the paired entry and exit rule lines, it is marked as a cross-border event (Crossing Event); recording information such as the time of event occurrence, trajectory path, involved camera ID, vehicle characteristics, etc.

[0094] Event display and storage specifically include: feeding back event display and storage to the management platform so that users can view cross-border event details in the management platform's historical event list; supporting click-through event playback trajectory animation and retrieving related video clips for replay analysis or evidence retention; in addition, event information can be pushed to third-party systems (such as traffic dispatch and security platforms) for coordinated response.

[0095] In combination with the first aspect, after step S124, the method further includes:

[0096] S125, intercepting a video clip from t seconds before the entry rule line is triggered to t seconds after the exit rule line is triggered.

[0097] S126, feeding back the cross-border event data, global trajectories, and video clips to the management end for storage.

[0098] Understandably, in order to provide more complete event context information and support subsequent analysis or evidence collection, the system will automatically capture video clips related to the event.

[0099] Where t seconds represents the time offset before the vehicle enters the zone's rule line. This is used to capture the vehicle's approach to the rule line, ensuring that the entire entry behavior is recorded. t seconds represents the time offset after the vehicle leaves the zone's rule line. This is used to record the vehicle's trajectory changes after leaving the rule line, ensuring that the entire cross-border process is fully captured.

[0100] Therefore, the intercepted time interval should be calculated as: (T_in-t, T_out-t2), where T_in is the time of entering the zone rule line and T_out is the time of leaving the zone rule line.

[0101] The system then extracts the video frames from the corresponding video streams for all cameras involved (including those at the entry and exit zones) within that time period and stitches them together into a complete video clip. If multiple cameras are involved in the global tracking, the system extracts the relevant video clips from each camera and packages them together for feedback.

[0102] As an implementable method, the system can overlay the following information on captured video clips to enhance visualization: global motion trajectory path, rule line position markers, vehicle target ID, color, model, and other metadata, as well as a timestamp watermark for accurate retrieval. Furthermore, to improve transmission efficiency and storage performance, the system encodes and compresses the video using common formats such as H.264 and H.265, while retaining sufficient clarity for subsequent viewing.

[0103] After completing the video clip capture, the system will uniformly feed back all data related to cross-border events to the management platform and store them in a structured manner for users to query, analyze and call.

[0104] In a second aspect, the present application also provides a vehicle cross-border tracking device, which is applied to a vehicle tracking system. The vehicle tracking system includes a management platform and an algorithm platform connected in sequence; Figure 5 As shown, the device includes: a configuration module 10 and a tracking module 20.

[0105] The configuration module 10 is used to configure the rule lines through the management platform and establish a logical pairing relationship between the entry rule line and the exit rule line to generate an algorithm task including the camera ID, the rule line coordinates and the area pairing relationship.

[0106] The tracking module 20 is used for the algorithm platform to receive algorithm tasks, and to perform cross-border vehicle tracking based on the configuration data of the algorithm tasks, and to feed back the tracking results to the management platform.

[0107] In a third aspect, the present application provides an electronic device, Figure 6 As shown, the electronic device includes a memory 131 and a processor 130. The memory 131 is used to store computer programs, and the processor 130 runs the computer programs to enable the electronic device to perform the above method.

[0108] Further, combined with Figure 6 The electronic device shown further includes a bus 132 and a communication interface 133 , and the processor 130 , the communication interface 133 and the memory 131 are connected via the bus 132 .

[0109] The memory 131 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk storage. The communication connection between the system network element and at least one other network element is achieved through at least one communication interface 133 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used. The bus 132 may be an ISA bus, a PCI bus, or an EISA bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 6 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0110] The processor 130 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by hardware integrated logic circuits in the processor 130 or by software instructions. The processor 130 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in conjunction with the embodiments of the present invention can be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in memory 131, and processor 130 reads information in memory 131 and, in conjunction with its hardware, completes the steps of the method of the aforementioned embodiment.

[0111] In a fourth aspect, an embodiment of the present application provides a readable storage medium, in which computer program instructions are stored. When the computer program instructions are read and executed by a processor, the above-mentioned method is executed.

[0112] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems and devices can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0113] In addition, in the description of the embodiments of the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; they may refer to mechanical connections or electrical connections; they may refer to direct connections or indirect connections through an intermediate medium; and they may refer to internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0114] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0115] In the description of the present invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limitations on the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0116] Finally, it should be noted that the above embodiments are only specific implementation methods of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that any person skilled in the art can modify or easily conceive of changes to the technical solutions described in the above embodiments within the technical scope disclosed by the present invention, or replace some of the technical features therein with equivalents. Such modifications, changes or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A vehicle cross-border tracking method, characterized in that: Applied to a vehicle tracking system, the vehicle tracking system includes a management platform and an algorithm platform connected in sequence; the method includes: By configuring rule lines on the management platform, and establishing a logical pairing relationship between entry rule lines and exit rule lines, an algorithm task is generated that includes camera ID, rule line coordinates, and area pairing relationship. The algorithm platform receives the algorithm task, performs cross-border vehicle tracking based on the configuration data of the algorithm task, and feeds back the tracking results to the management platform.

2. The method according to claim 1, characterized in that The management platform provides a graphical display interface; the steps of configuring monitoring rules through the management platform include: In response to the addition operation for monitoring, the video stream corresponding to the added monitoring device code is displayed on the bottom canvas interactive interface; In response to a configuration operation for a monitoring rule, floatingly displaying an upper transparent canvas interaction interface on the bottom canvas interaction interface; In response to click operations on multiple points in the upper transparent canvas interactive interface, each point is used as a regular point and connected in sequence according to the click sequence to form a closed polygon; The closed polygon is used as a monitoring area, the vehicles in the monitoring area are used as vehicles to be monitored, and the code of the monitoring device corresponding to the monitoring area is used as the monitoring device code to obtain a monitoring rule.

3. The method according to claim 2, characterized in that The steps to configure monitoring rules through the management platform include: In response to the triggering operation of the first designated key, all the rule points and the currently constructed monitoring area in the upper transparent canvas interaction interface are deleted.

4. The method according to claim 3, characterized in that After configuring monitoring rules through the management platform, the following steps are also included: In response to the drag operation on the target rule point, the current position of the target rule point, the monitoring area, and the monitoring device code are updated to obtain an updated monitoring rule.

5. The method according to claim 1, wherein The steps for establishing a logical pairing relationship between an entry rule line and an exit rule line include: In response to the area association instruction, the entry rule line and the exit rule line are bound into a pairing group; A unique region identifier is set for the pairing group and associated with the camera ID.

6. The method according to claim 1, characterized in that The steps of performing cross-border vehicle tracking based on the configuration data of the algorithm task include: Using a deep learning detection model to detect vehicles in real time in the video frames corresponding to the configuration rule lines, and generate vehicle bounding box information with timestamps; Based on the vehicle bounding box information, a tracking algorithm is used to associate vehicle targets in consecutive video frames to generate a vehicle motion trajectory within a single camera; Match vehicle trajectories with depth feature similarity above a threshold and spatiotemporal continuity from different cameras and fit the global motion trajectory through coordinate transformation; When the global trajectory triggers the paired entry rule line and exit rule line in sequence, it is marked as a cross-border event.

7. The method according to claim 6, characterized in that When the global trajectory triggers the paired entry rule line and exit rule line in sequence, after marking it as a cross-border event, the method further includes: Intercept the video clip from t seconds before the entry rule line is triggered to t seconds after the exit rule line is triggered; Feedback cross-border event data, global trajectories, and video clips to the management end for storage.

8. A vehicle cross-border tracking device, characterized in that: Applied to a vehicle tracking system, the vehicle tracking system includes a management platform and an algorithm platform connected in sequence; the device includes: The configuration module is used to configure the rule lines through the management platform and establish a logical pairing relationship between the entry rule line and the exit rule line to generate an algorithm task containing the camera ID, rule line coordinates and area pairing relationship; The tracking module is used for the algorithm platform to receive the algorithm task, perform cross-border vehicle tracking based on the configuration data of the algorithm task, and feed back the tracking results to the management platform.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to perform the method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores computer program instructions, and when the computer program instructions are read and executed by a processor, the method according to any one of claims 1 to 7 is executed.

Images