Suspicious vehicle path tracking method and system based on spatio-temporal reverse search

Abstract

The application provides a suspicious vehicle path tracking method and system based on spatiotemporal reverse search. The method comprises the following steps: acquiring road network structure information related to a preset toll station and vehicle feature library information of a target suspicious vehicle, extracting path branches and matching path priorities according to the road network structure information, acquiring road section scene data corresponding to each path branch, processing to obtain vehicle speed probability density and a corrected time window, acquiring video data, processing to obtain vehicle matching degrees in combination with the vehicle feature library information, and obtaining a final complete path in combination with the vehicle matching degrees according to the corrected time window, so as to realize the suspicious vehicle path tracking technology based on spatiotemporal reverse search.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of path tracking, in particular to a suspicious vehicle path tracking method and system based on spatiotemporal reverse search. BACKGROUND

[0002] In the process of traffic management and case investigation, path tracking of suspicious vehicles is crucial. Traditional vehicle path tracking methods often rely on manual video monitoring, which is inefficient and prone to omissions or misjudgments due to human factors. With the development of intelligent transportation systems, some tracking methods based on video analysis and road network information have emerged. However, these methods struggle to accurately extract effective path information in the face of complex road networks and massive data, and are not precise enough in matching vehicle features and time windows, resulting in unsatisfactory tracking results.

[0003] Existing path tracking techniques mostly track from the starting point to the ending point. In cases where the suspicious vehicle's driving trajectory is unclear, it can easily fall into the dilemma of excessive path branching, increasing the difficulty and workload of tracking. The method based on spatiotemporal reverse search can effectively reduce the tracking range and improve the tracking efficiency by tracking from known key nodes such as toll stations. However, current related technologies have deficiencies in path branch priority matching, time window correction, and vehicle matching degree combination, which need further optimization.

[0004] In view of the above problems, an effective technical solution is urgently needed. SUMMARY

[0005] The purpose of the present application is to provide a suspicious vehicle path tracking method and system based on spatiotemporal reverse search, which can obtain road network structure information related to a preset toll station and vehicle feature library information of a target suspicious vehicle, extract path branches and match path priorities based on the road network structure information, obtain road segment scene data corresponding to each path branch, process vehicle speed probability density and corrected time windows, obtain video data, process vehicle matching degrees in combination with the vehicle feature library information, and obtain the final complete path based on the corrected time windows and vehicle matching degrees, thereby realizing the suspicious vehicle path tracking technology based on spatiotemporal reverse search.

[0006] The present application also provides a suspicious vehicle path tracking method based on spatiotemporal reverse search, comprising the following steps:

[0007] Obtain road network structure information related to a preset toll station and vehicle feature library information of a target suspicious vehicle;

[0008] Extract path branches and match path priorities based on the road network structure information;

[0009] obtain the road section scene data corresponding to each path branch, and process to obtain the vehicle speed probability density and the corrected time window;

[0010] obtain the video data, and process to obtain the vehicle matching degree in combination with the vehicle feature library information;

[0011] obtain the final complete path in combination with the vehicle matching degree according to the corrected time window.

[0012] Optionally, in the suspicious vehicle path tracking method based on spatiotemporal reverse retrieval provided in the present application, the obtaining of the road network structure information related to the preset toll station and the vehicle feature library information of the target suspicious vehicle comprises:

[0013] if the target suspicious vehicle is found at the preset toll station, obtaining the road network structure information related to the preset toll station and the vehicle feature library information of the target suspicious vehicle;

[0014] the road network structure information comprises the terminal position, the terminal time, the road section length, the entry information and the connection relationship information;

[0015] the vehicle feature library information comprises the license plate number, the vehicle model, the color, the vehicle hub and the tire model.

[0016] Optionally, in the suspicious vehicle path tracking method based on spatiotemporal reverse retrieval provided in the present application, the extraction of the path branches and the matching of the path priorities according to the road network structure information comprises:

[0017] processing according to the terminal position, the terminal time, the road section length, the entry information and the connection relationship information by using a preset path prediction model to obtain each path branch;

[0018] obtaining the historical driving data of the target suspicious vehicle, comprising the branch selection record and the historical driving frequency corresponding to each path branch;

[0019] performing priority sorting on each path branch according to the historical driving data to obtain the priority corresponding to each path branch.

[0020] Optionally, in the suspicious vehicle path tracking method based on spatiotemporal reverse retrieval provided in the present application, the obtaining of the road section scene data corresponding to each path branch and the processing to obtain the vehicle speed probability density and the corrected time window comprise:

[0021] obtaining the road section scene data corresponding to each path branch, comprising the branch road section length, the road section historical vehicle speed distribution, the real-time vehicle flow speed data and the real-time weather data;

[0022] obtaining the vehicle speed probability density according to the road section historical vehicle speed distribution;

[0023] According to the branch road length, vehicle speed probability density and end point time, a time window is obtained;

[0024] According to the real-time vehicle flow speed data and real-time weather data, the time window is corrected to obtain a corrected time window.

[0025] Optionally, in the suspicious vehicle path tracking method based on space-time reverse retrieval, the video data is obtained, and vehicle matching degrees are obtained by processing the vehicle feature library information.

[0026] Video data of each path branch monitoring point is obtained, and vehicle feature library information of a vehicle to be verified is extracted;

[0027] According to the vehicle feature library information of the vehicle to be verified and the vehicle feature library information, corresponding feature matching is performed, and weighted processing is performed to obtain a vehicle matching degree.

[0028] Optionally, in the suspicious vehicle path tracking method based on space-time reverse retrieval, the corrected time window is combined with the vehicle matching degree to obtain a final complete path.

[0029] According to the vehicle matching degree and a preset vehicle matching threshold, threshold comparison is performed.

[0030] If the vehicle matching degree is greater than the preset vehicle matching threshold, the vehicle matching degree is marked as a target matching degree, and a corresponding target monitoring point position and time are obtained.

[0031] According to the corrected time window, the target monitoring point position and the time are combined to generate a final complete path.

[0032] In a second aspect, the present application provides a suspicious vehicle path tracking system based on space-time reverse retrieval, which comprises a memory and a processor, wherein the memory comprises a program of the suspicious vehicle path tracking method based on space-time reverse retrieval, and the program of the suspicious vehicle path tracking method based on space-time reverse retrieval is executed by the processor to realize the following steps:

[0033] Obtain road network structure information related to a preset toll station and vehicle feature library information of a target suspicious vehicle;

[0034] According to the road network structure information, path branches are extracted and path priorities are matched;

[0035] Obtain road section scene data corresponding to each path branch, and process to obtain vehicle speed probability density and a corrected time window;

[0036] Obtain video data, and process the vehicle feature library information to obtain vehicle matching degrees;

[0037] According to the modified time window, a final complete path is obtained in combination with the vehicle matching degree.

[0038] Optionally, in the suspicious vehicle path tracking system based on spatio-temporal reverse retrieval, the acquisition of the road network structure information related to the preset toll station and the vehicle feature library information of the target suspicious vehicle comprises:

[0039] If the target suspicious vehicle is found at the preset toll station, the road network structure information related to the preset toll station and the vehicle feature library information of the target suspicious vehicle are acquired;

[0040] The road network structure information comprises a terminal position, a terminal time, a road section length, entry information and connection relationship information;

[0041] The vehicle feature library information comprises a license plate number, a vehicle model, a color, a vehicle hub and a tire model.

[0042] Optionally, in the suspicious vehicle path tracking system based on spatio-temporal reverse retrieval, the extraction of path branches and the matching of path priorities according to the road network structure information comprises:

[0043] According to the terminal position, the terminal time, the road section length, the entry information and the connection relationship information, each path branch is obtained by processing through a preset path prediction model;

[0044] The historical driving data of the target suspicious vehicle are acquired, comprising branch selection records and historical driving frequencies corresponding to each path branch;

[0045] According to the historical driving data, each path branch is prioritized to obtain a priority corresponding to each path branch.

[0046] Optionally, in the suspicious vehicle path tracking system based on spatio-temporal reverse retrieval, the acquisition of the road section scene data corresponding to each path branch, the processing to obtain a vehicle speed probability density and a modified time window, comprises:

[0047] The road section scene data corresponding to each path branch are acquired, comprising a branch road section length, a road section historical vehicle speed distribution, real-time traffic flow speed data and real-time weather data;

[0048] According to the road section historical vehicle speed distribution, a vehicle speed probability density is acquired;

[0049] According to the branch road section length, the vehicle speed probability density and the terminal time, a time window is obtained by processing;

[0050] According to the real-time traffic flow speed data and the real-time weather data, the time window is modified to obtain a modified time window.

[0051] From the above, the suspicious vehicle path tracking method and system based on spatiotemporal reverse retrieval provided by the application, by acquiring the road network structure information related to the preset toll station and the vehicle feature library information of the target suspicious vehicle, extracting the path branches according to the road network structure information and matching the path priority, acquiring the road section scene data corresponding to each path branch, processing to obtain the vehicle speed probability density and the corrected time window, acquiring the video data, processing to obtain the vehicle matching degree in combination with the vehicle feature library information, and obtaining the final complete path according to the corrected time window in combination with the vehicle matching degree, so as to realize the suspicious vehicle path tracking technology based on spatiotemporal reverse retrieval.

[0052] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application can be realized and achieved by the structure particularly pointed out in the written description and the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0053] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments of the present application. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be considered as limiting the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0054] Figure 1 The flowchart of the suspicious vehicle path tracking method based on spatiotemporal reverse retrieval provided by the embodiments of the present application;

[0055] Figure 2 The flowchart of matching path priority of the suspicious vehicle path tracking method based on spatiotemporal reverse retrieval provided by the embodiments of the present application;

[0056] Figure 3 The flowchart of obtaining vehicle speed probability density and corrected time window of the suspicious vehicle path tracking method based on spatiotemporal reverse retrieval provided by the embodiments of the present application;

[0057] Figure 4 The flowchart of obtaining vehicle matching degree of the suspicious vehicle path tracking method based on spatiotemporal reverse retrieval provided by the embodiments of the present application. DETAILED DESCRIPTION

[0058] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0059] It should be noted that similar reference numbers and letters represent similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0060] Please refer to Figure 1 , Figure 1 is a flowchart of a suspicious vehicle path tracking method based on spatio-temporal reverse retrieval in some embodiments of the present application. The suspicious vehicle path tracking method based on spatio-temporal reverse retrieval is used in a terminal device, such as a computer, a mobile phone terminal, etc. The suspicious vehicle path tracking method based on spatio-temporal reverse retrieval includes the following steps:

[0061] S11, obtaining road network structure information related to a preset toll station and vehicle feature library information of a target suspicious vehicle;

[0062] S12, extracting path branches and matching path priorities according to the road network structure information;

[0063] S13, obtaining road section scene data corresponding to each path branch, processing to obtain vehicle speed probability density and a corrected time window;

[0064] S14, obtaining video data, and processing to obtain vehicle matching degree in combination with the vehicle feature library information;

[0065] S15, obtaining a final complete path in combination with vehicle matching degree according to the corrected time window.

[0066] It needs to be explained that the suspicious vehicle A found at the exit of a toll station on a highway needs to determine the driving path of the vehicle A because there are multiple paths in the road network and multiple toll stations to reach the toll station, and the video data to be processed is massive by using the ordinary video retrieval method, therefore, in order to solve these problems, first, the road network structure information related to the preset toll station and the vehicle feature library information of the target suspicious vehicle are acquired, wherein the road network structure information includes the terminal position, the terminal time, the road section length, the entry information and the connection relationship information, and the vehicle feature library information includes the license plate number, the vehicle type, the color, the vehicle hub and the tire model, then, the path branch is extracted according to the road network structure information and the path priority is matched, the road section scene data corresponding to each path branch is acquired, including the branch road section length, the historical vehicle speed distribution of the road section, the real-time vehicle flow speed data and the real-time weather data, the vehicle speed probability density and the corrected time window are obtained by processing, the video data is acquired, the vehicle matching degree is obtained by processing in combination with the vehicle feature library information, the final complete path is obtained according to the corrected time window in combination with the vehicle matching degree, so that the suspicious vehicle path tracking technology based on the spatiotemporal reverse retrieval is realized.

[0067] According to the embodiment of the application, the road network structure information related to the preset toll station and the vehicle feature library information of the target suspicious vehicle are acquired, including:

[0068] If the target suspicious vehicle is found at the preset toll station, the road network structure information related to the preset toll station and the vehicle feature library information of the target suspicious vehicle are acquired;

[0069] The road network structure information includes the terminal position, the terminal time, the road section length, the entry information and the connection relationship information;

[0070] The vehicle feature library information includes the license plate number, the vehicle type, the color, the vehicle hub and the tire model.

[0071] It should be noted that when the target suspicious vehicle is monitored at the preset toll station, the information collection mechanism needs to be started in the first time, and two types of key data are comprehensively acquired: the road network structure information associated with the preset toll station and the feature library information of the target suspicious vehicle, to provide accurate basis for subsequent tracking and interception. The road network structure information covers multiple dimensions of core elements: the specific position of the vehicle when it is found at the end position, which defines the boundary of the control range; the end time is the specific time when the suspicious vehicle passes through the toll station, which can be combined with the driving path to calculate the reasonable travel time, and assist in judging whether the vehicle has abnormal stay or detour; the road section length and road grade are directly related to the traffic efficiency, which can estimate the driving time of the vehicle on each road section; the entry information marks all possible branch intersections along the way, avoiding the omission of the path in the tracking process; the connection relationship information clearly presents the connection logic of each road section, providing topological basis for analyzing the possible lane changing and turning routes of the vehicle; and in the vehicle feature library information, the license plate number is the primary identifier for identifying the vehicle, which can quickly associate the vehicle registration information; the vehicle model (such as sedan, truck, SUV, etc.) determines its traffic restriction and road selection preference; the vehicle color is convenient for quick positioning in complex traffic flow; the style, size and tire model of the vehicle hub serve as unique details, which can provide auxiliary identification basis when the license plate is obscured or tampered with, greatly improving the tracking recognition degree.

[0072] Please refer to Figure 2 , Figure 2 is a flowchart of matching path priority of a suspicious vehicle path tracking method based on spatio-temporal reverse retrieval in some embodiments of the present application. According to the embodiment of the present application, the path branches are extracted according to the road network structure information, and the path priority is matched, which includes:

[0073] S21, processing according to the end position, end time, road section length, entry information and connection relationship information through a preset path prediction model to obtain each path branch;

[0074] S22, obtaining the historical driving data of the target suspicious vehicle, including branch selection records and historical driving frequencies corresponding to each path branch;

[0075] S23, prioritizing each path branch according to the historical driving data to obtain the priority corresponding to each path branch.

[0076] It should be noted that after obtaining the road network structure information of the preset toll station, the terminal position, the terminal time, the road section length, the information of the merging entrance and the connection relationship information are comprehensively processed by means of the preset path prediction model. The model can generate all possible path branches from the current position to the terminal by analyzing the spatial correlation and traffic constraints of each road section, covering the main road, auxiliary road and alternative routes derived from each merging entrance, thereby providing a complete path reference framework for subsequent tracking. At the same time, the historical driving data of the target suspicious vehicle need to be called, in which the core information includes branch selection records and historical driving frequencies corresponding to each path branch. The branch selection records can intuitively present the vehicle's past preference for different forks in a similar road network structure, for example, whether to turn left or straight at a specific merging entrance. The historical driving frequency reflects the degree of dependence of the vehicle on a certain path through quantitative data, such as the number of times of selecting a certain provincial road accounting for 60% between the same start and end points. Based on the above historical data, the generated path branches can be prioritized. The sorting logic takes the historical selection rule as the core, marks the path frequently selected by the vehicle in the past as high priority, sets the path occasionally selected as medium priority, and lists the path never selected as low priority. Through this sorting, the tracking efficiency can be greatly improved, and the law enforcement resources can be focused on the most likely selected path.

[0077] Please refer to Figure 3 , Figure 3 is a flowchart of obtaining vehicle speed probability density and a corrected time window in the suspicious vehicle path tracking method based on spatio-temporal reverse retrieval in some embodiments of the present application. According to the embodiment of the present application, the acquisition of the road section scene data corresponding to each path branch, the processing of the obtained vehicle speed probability density and the corrected time window, comprises:

[0078] S31, acquiring the road section scene data corresponding to each path branch, including branch road section length, road section historical vehicle speed distribution, real-time traffic speed data and real-time weather data;

[0079] S32, obtaining vehicle speed probability density according to the road section historical vehicle speed distribution;

[0080] S33, processing according to the branch road section length, vehicle speed probability density and terminal time to obtain a time window;

[0081] S34, correcting the time window according to the real-time traffic speed data and real-time weather data to obtain a corrected time window.

[0082] It should be noted that, in order to accurately determine the passing time of the target suspicious vehicle at each path branch, the corresponding road section scene data needs to be obtained, specifically including branch road section length, road section historical vehicle speed distribution, real-time vehicle flow speed data and real-time weather data; based on the road section historical vehicle speed distribution, the vehicle speed probability density can be obtained through statistical analysis, which can reflect the possibility of different vehicle speeds appearing on the road section, for example, in a certain suburban road section, the appearance probability of vehicle speed of 60-80 kilometers / hour is 70%, which provides a probability basis for predicting the passing time; then, combined with the branch road section length, the vehicle speed probability density and the end time, the time window is obtained. Specifically, by performing division operation on the road section length and the vehicle speed under different probabilities, a plurality of possible passing times are obtained, and then combined with the end time to back calculate the time range of the vehicle passing through the road section, to form an initial time window. Since real-time road conditions and weather can significantly affect the passing efficiency, real-time vehicle flow speed data and real-time weather data are needed to correct the initial time window; for example, when the real-time vehicle flow speed is lower than the historical average by 30%, the time window needs to be extended; in the case of heavy rain, the time window needs to be further expanded, for example, the initial time window is set to 30-40 minutes, if heavy rain occurs, the window needs to be adjusted to 45-60 minutes according to the historical rainfall day vehicle speed decay model; if the visibility is less than 50 meters in heavy fog, an elastic interval needs to be added in the window, marked as “may be delayed for 15-30 minutes”; during the correction process, a two-factor weight model needs to be established: when the real-time vehicle flow and weather data have the same effect (such as congestion + heavy rain), the window is extended; when the data have opposite effects (such as smooth + light rain), the vehicle flow speed is dominant, and the window range is fine-tuned, to finally obtain a corrected time window that is more in line with the actual situation, providing a time basis for accurate control.

[0083] Please refer to Figure 4 , Figure 4 is a flowchart of obtaining a vehicle matching degree in a suspicious vehicle path tracking method based on spatiotemporal reverse retrieval in some embodiments of the present application. According to the embodiment of the present application, the video data is obtained, and the vehicle matching degree is obtained by processing the vehicle feature library information, which comprises:

[0084] S41, obtaining video data of each path branch monitoring point, and extracting the to-be-verified vehicle feature library information;

[0085] S42, performing corresponding feature matching on the to-be-verified vehicle feature library information and the vehicle feature library information respectively, and performing weighted processing to obtain a vehicle matching degree.

[0086] It should be noted that the system collects video data of each path branch monitoring point in order of priority of path branches, extracts feature information of the vehicle to be verified by means of image recognition technology, and specifically includes license plate number, vehicle model, color, vehicle hub and tire model and other parameters; then, the feature library of the vehicle to be verified and the feature library of the target suspicious vehicle are matched in each dimension, wherein the license plate number is accurately compared by characters, the vehicle model is matched by contour and parameter library, the color is calibrated by spectrum analysis, and the hub and tire model are identified by combining pattern texture and size data; after matching, weights are assigned according to feature importance, for example, the license plate number accounts for 0.3, the vehicle model and the color each account for 0.2, and the hub and the tire model account for 0.3 together; the vehicle matching degree in the range of 0-100 is generated by weighted summation algorithm, and accurate quantitative comparison of multiple feature dimensions is realized.

[0087] According to the embodiment of the present application, the final complete path is obtained according to the modified time window combined with the vehicle matching degree, comprising:

[0088] The vehicle matching degree is compared with a preset vehicle matching threshold value;

[0089] If the vehicle matching degree is greater than the preset vehicle matching threshold value, the vehicle matching degree is marked as a target matching degree, and the corresponding target monitoring point position and time are obtained;

[0090] The final complete path is generated according to the modified time window combined with the target monitoring point position and time.

[0091] It should be noted that the system compares each vehicle matching degree calculated by the system with a preset vehicle matching threshold value one by one for accurate comparison; this threshold value is set based on historical data and recognition accuracy requirements, and is used to filter vehicle information that meets the basic matching standard; when the matching degree of a vehicle exceeds the preset threshold value, the system will automatically mark it as a target matching degree, and quickly extract the target monitoring point position information corresponding to the matching result and the time data of the vehicle passing through the monitoring point; these information will be temporarily stored as key node data for path generation; then, the system calls the modified time window parameters, combines the target monitoring point position and time obtained before, and the system connects each node in time sequence by time sequence analysis and spatial correlation algorithm, and finally generates a complete and continuous driving path of the target suspicious vehicle.

[0092] In a second aspect, the present application also discloses a suspicious vehicle path tracking system based on spatiotemporal reverse search, which comprises a memory and a processor, the memory comprises a suspicious vehicle path tracking method program based on spatiotemporal reverse search, and the suspicious vehicle path tracking method program based on spatiotemporal reverse search is implemented when the processor is executed.

[0093] Obtain road network structure information related to a preset toll station and vehicle feature library information of a target suspicious vehicle;

[0094] Extract path branches according to the road network structure information and match path priorities;

[0095] Obtain road section scene data corresponding to each path branch, process to obtain vehicle speed probability density and a corrected time window;

[0096] Obtain video data, process to obtain vehicle matching degree in combination with the vehicle feature library information;

[0097] Obtain a final complete path according to the corrected time window in combination with the vehicle matching degree.

[0098] It should be noted that a suspicious vehicle A found at an exit of a toll station on an expressway needs to determine the driving path of the vehicle A because there are multiple paths and multiple toll stations reaching the toll station in the road network. The video data to be processed is massive by using an ordinary video retrieval method. Therefore, in order to solve these problems, first, obtain road network structure information related to a preset toll station and vehicle feature library information of a target suspicious vehicle, wherein the road network structure information includes terminal position, terminal time, road section length, entry information and connection relationship information, and the vehicle feature library information includes license plate number, vehicle type, color, vehicle hub and tire model. Then, extract path branches according to the road network structure information and match path priorities, obtain road section scene data corresponding to each path branch, including branch road section length, road section historical vehicle speed distribution, real-time vehicle flow speed data and real-time weather data, process to obtain vehicle speed probability density and a corrected time window, obtain video data, process to obtain vehicle matching degree in combination with the vehicle feature library information, and obtain a final complete path according to the corrected time window in combination with the vehicle matching degree, so as to realize the technology of suspicious vehicle path tracking based on spatiotemporal reverse retrieval.

[0099] According to an embodiment of the present application, the obtaining of the road network structure information related to the preset toll station and the vehicle feature library information of the target suspicious vehicle includes:

[0100] If a target suspicious vehicle is found at a preset toll station, obtain road network structure information related to the preset toll station and vehicle feature library information of the target suspicious vehicle;

[0101] The road network structure information includes terminal position, terminal time, road section length, entry information and connection relationship information;

[0102] The vehicle feature library information includes license plate number, vehicle type, color, vehicle hub and tire model.

[0103] It should be noted that when the target suspicious vehicle is monitored at the preset toll station, the information collection mechanism needs to be started in the first time, and two types of key data are comprehensively acquired: the road network structure information associated with the preset toll station and the feature library information of the target suspicious vehicle, to provide accurate basis for subsequent tracking and interception, wherein the road network structure information covers multi-dimensional core elements: the specific position of the vehicle when the end position is clear, which defines the boundary of the control range; the end time is the specific time when the suspicious vehicle passes through the toll station, which can be combined with the driving path to calculate the reasonable travel time, to assist in judging whether the vehicle exists abnormal stay or detour; the road section length and the road grade are directly related to the traffic efficiency, which can estimate the driving time of the vehicle in each road section; the entry information marks all possible branch intersections along the way, to avoid missing the path in the tracking process; the connection relationship information clearly presents the connection logic of each road section, to provide topological basis for analyzing the possible lane changing and turning route of the vehicle; and in the vehicle feature library information, the license plate number is the primary identification of the vehicle, which can be quickly associated with the vehicle registration information; the vehicle model (such as car, truck, SUV, etc.) determines its traffic restriction and road selection preference; the vehicle body color is convenient for quick positioning in complex traffic flow; the style, size and tire model of the vehicle hub serve as unique details, which can provide auxiliary identification basis when the license plate is blocked or tampered, and greatly improve the tracking recognition degree.

[0104] According to the embodiment of the application, the path branches are extracted and the path priority is matched according to the road network structure information, comprising:

[0105] According to the end position, the end time, the road section length, the entry information and the connection relationship information, the path branches are obtained by processing through a preset path prediction model.

[0106] The historical driving data of the target suspicious vehicle are acquired, including branch selection records and historical driving frequencies corresponding to each path branch.

[0107] According to the historical driving data, the priority of each path branch is sorted, and the priority corresponding to each path branch is obtained.

[0108] It should be noted that after obtaining the road network structure information of the preset toll station, the terminal position, the terminal time, the road section length, the information of the merging entrance and the connection relationship information are comprehensively processed by means of the preset path prediction model. The model can generate all possible path branches from the current position to the terminal by analyzing the spatial correlation and traffic constraints of each road section, covering the main road, auxiliary road and the alternative routes derived from each merging entrance, thereby providing a complete path reference framework for subsequent tracking. At the same time, the historical driving data of the target suspicious vehicle need to be called, in which the core information includes branch selection records and historical driving frequencies corresponding to each path branch. The branch selection records can intuitively present the vehicle's past preference for different forks in a similar road network structure, for example, whether to turn left or straight at a specific merging entrance. The historical driving frequency reflects the degree of dependence of the vehicle on a certain path through quantitative data, such as the number of times of selecting a certain provincial road accounting for 60% between the same start and end points. Based on the above historical data, the generated path branches can be prioritized. The sorting logic takes the historical selection rule as the core, marks the path frequently selected by the vehicle in the past as high priority, sets the path occasionally selected as medium priority, and lists the path never selected as low priority. Through this sorting, the tracking efficiency can be greatly improved, and the law enforcement resources can be preferentially focused on the most likely selected path.

[0109] According to the embodiment of the application, the obtaining of the road section scene data corresponding to each path branch, the processing of the obtained vehicle speed probability density and the corrected time window, comprises:

[0110] The road section scene data corresponding to each path branch includes branch road section length, road section historical vehicle speed distribution, real-time traffic speed data and real-time weather data.

[0111] The vehicle speed probability density is obtained according to the road section historical vehicle speed distribution.

[0112] The time window is obtained by processing the branch road section length, the vehicle speed probability density and the terminal time.

[0113] The time window is corrected according to the real-time traffic speed data and the real-time weather data to obtain the corrected time window.

[0114] It should be noted that, in order to accurately determine the passing time of the target suspicious vehicle at each path branch, the corresponding road section scene data needs to be obtained, specifically including branch road section length, road section historical vehicle speed distribution, real-time vehicle flow speed data and real-time weather data; based on the road section historical vehicle speed distribution, the vehicle speed probability density can be obtained through statistical analysis, and this data can reflect the possibility of different vehicle speeds appearing on the road section, for example, in a certain suburban road section, the appearance probability of vehicle speed of 60-80 kilometers / hour is 70%, which provides a probability basis for predicting the passing time; then, combined with the branch road section length, the vehicle speed probability density and the end time, the time window is calculated and processed. Specifically, by performing division operation on the road section length and the vehicle speed under different probabilities, a plurality of possible passing time lengths are obtained, and then combined with the end time to back-propagate, the time range of the vehicle passing through the road section is framed to form an initial time window. Since real-time road conditions and weather can significantly affect the passing efficiency, the real-time vehicle flow speed data and the real-time weather data need to be used to correct the initial time window; for example, when the real-time vehicle flow speed is lower than the historical average by 30%, the time window needs to be extended; in the case of heavy rain weather, the time range is further expanded, for example, the initial time window is set to 30-40 minutes, if heavy rain suddenly occurs, the window needs to be adjusted to 45-60 minutes according to the historical rainfall day vehicle speed attenuation model; if the visibility is less than 50 meters in heavy fog, an elastic interval needs to be added in the window, and "may be delayed for 15-30 minutes" is marked; during the correction process, a two-factor weight model needs to be established: when the real-time vehicle flow and weather data have the same effect (such as congestion + heavy rain), the window is extended; when the data acts in the opposite direction (such as smooth + light rain), the vehicle flow speed is dominant, and the window range is fine-tuned, and finally a more actual corrected time window is obtained, which provides a time basis for accurate control.

[0115] According to the embodiment of the application, the video data is obtained, and the vehicle matching degree is obtained by processing the vehicle feature library information.

[0116] Video data of each path branch monitoring point is obtained, and vehicle feature library information to be verified is extracted;

[0117] The vehicle matching degree is obtained by performing corresponding feature matching on the vehicle feature library information to be verified and the vehicle feature library information respectively and performing weighted processing.

[0118] It should be noted that the system collects the video data of each path branch monitoring point in the order of priority of the path branches, extracts the feature information of the vehicle to be verified by means of image recognition technology, and specifically includes the license plate number, vehicle model, color, vehicle hub and tire model and other parameters; then, the feature library of the vehicle to be verified and the feature library of the target suspicious vehicle are matched in each dimension, wherein the license plate number is accurately compared by characters, the vehicle model is matched by contour and parameter library, the color is calibrated by spectrum analysis, and the hub and tire model are identified by combining pattern texture and size data; after the matching is completed, the weight is distributed according to the importance of the features, for example, the license plate number accounts for 0.3, the vehicle model and the color each account for 0.2, and the hub and the tire model account for 0.3 together; the vehicle matching degree in the range of 0-100 is generated by weighted summation algorithm, and the accurate quantitative comparison of multiple feature dimensions is realized.

[0119] According to the embodiment of the application, the final complete path is obtained according to the modified time window and the vehicle matching degree, comprising:

[0120] According to the vehicle matching degree and the preset vehicle matching threshold value, the threshold comparison is performed;

[0121] If the vehicle matching degree is greater than the preset vehicle matching threshold value, the vehicle matching degree is marked as a target matching degree, and the corresponding target monitoring point position and time are obtained;

[0122] According to the modified time window, the target monitoring point position and the time are generated to form a final complete path.

[0123] It should be noted that the system compares each vehicle matching degree calculated by the system with the preset vehicle matching threshold value one by one; this threshold value is set based on historical data and recognition accuracy requirements, and is used to filter vehicle information that meets the basic matching standard; when the matching degree of a vehicle exceeds the preset threshold value, the system will automatically mark it as a target matching degree, and quickly extract the target monitoring point position information corresponding to the matching result and the time data of the vehicle passing through the monitoring point; these information will be temporarily stored as key node data for path generation; then, the system calls the modified time window parameter, combines the target monitoring point position and time obtained before, and connects each node in time sequence by time sequence analysis and spatial correlation algorithm, to finally generate a complete and continuous driving path of the target suspicious vehicle.

[0124] The suspicious vehicle path tracking method and system based on space-time reverse search disclosed by the application realizes the suspicious vehicle path tracking based on space-time reverse search by acquiring the road network structure information related to the preset toll station and the vehicle feature library information of the target suspicious vehicle, extracting path branches according to the road network structure information and matching path priorities, acquiring the road section scene data corresponding to each path branch, processing the vehicle speed probability density and the corrected time window, acquiring the video data, processing the vehicle matching degree in combination with the vehicle feature library information, and acquiring the final complete path according to the corrected time window in combination with the vehicle matching degree.

[0125] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other manners. The above described device embodiments are merely schematic, for example, the division of the units is only a logical function division, and there can be another division manner in actual implementation, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the above display or discussed components can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0126] The units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units; they can be located in one place, or distributed on multiple network units; and part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0127] In addition, each functional unit in each embodiment of the application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be realized in the form of hardware, or in the form of hardware plus software functional unit.

[0128] Those of ordinary skill in the art can understand that all or part of the steps of the above method embodiments can be completed by program instruction related hardware, and the foregoing program can be stored in a readable storage medium, and the program is executed to perform the steps of the above method embodiments; and the foregoing storage medium includes mobile storage devices, read-only memories (ROMs), random access memories (RAMs), magnetic discs or optical discs, and various storage media that can store program codes.

[0129] Alternatively, the above-mentioned integrated unit of the present application, if realized in the form of a software function module and sold or used as an independent product, can also be stored in a readable storage medium. Based on such understanding, the technical solutions of the embodiments of the present application can be embodied in the form of a software product, which is stored in a storage medium, includes several instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: mobile storage devices, ROM, RAM, magnetic or optical disks, and various media that can store program codes.

Claims

1. A method for tracing the path of suspicious vehicles based on spatiotemporal reverse retrieval, characterized in that, Includes the following steps: Obtain road network structure information related to preset toll stations and vehicle feature database information of target suspicious vehicles; Extract path branches and match path priorities based on the road network structure information; Obtain the road segment scene data corresponding to each path branch, process it to obtain the vehicle speed probability density and the corrected time window; Acquire video data and process it in conjunction with the vehicle feature database information to obtain the vehicle matching degree; The final complete path is obtained based on the corrected time window and the vehicle matching degree. The acquisition of road network structure information related to preset toll stations and vehicle feature database information of target suspicious vehicles includes: If a suspicious vehicle is detected at a pre-set toll station, obtain the relevant road network structure information of the pre-set toll station and the vehicle feature database information of the suspicious vehicle. The road network structure information includes the destination location, destination time, road segment length, merging point information, and connection relationship information; The vehicle feature database information includes license plate number, vehicle model, color, vehicle wheel rims, and tire model; The step of extracting path branches and matching path priorities based on the road network structure information includes: The path branches are obtained by processing the endpoint location, endpoint time, road segment length, merging point information, and connection relationship information through a preset path prediction model. Obtain the historical driving data of the target suspicious vehicle, including branch selection records and the historical driving frequency corresponding to each path branch; Based on the historical driving data, the priority of each path branch is sorted to obtain the priority of each path branch; The process of acquiring road segment scene data corresponding to each path branch, and processing it to obtain vehicle speed probability density and corrected time windows includes: Obtain the road segment scene data corresponding to each path branch, including branch road segment length, historical vehicle speed distribution of the road segment, real-time traffic flow speed data, and real-time weather data; The vehicle speed probability density is obtained based on the historical vehicle speed distribution of the road segment. The time window is obtained by processing the branch segment length, vehicle speed probability density, and destination time. The time window is corrected based on the real-time traffic speed data and real-time weather data to obtain the corrected time window.

2. The method for tracing the path of suspicious vehicles based on spatiotemporal reverse retrieval according to claim 1, characterized in that, The process of acquiring video data and combining it with the vehicle feature database information to obtain vehicle matching degree includes: Acquire video data from monitoring points at each path branch and extract feature database information of the vehicle to be verified. Based on the vehicle feature library information to be verified and the vehicle feature library information, corresponding feature matching is performed respectively, and weighted processing is performed to obtain the vehicle matching degree.

3. The method for tracing the path of suspicious vehicles based on spatiotemporal reverse retrieval according to claim 2, characterized in that, The step of obtaining the final complete path based on the corrected time window and vehicle matching degree includes: The threshold is compared with the preset vehicle matching threshold based on the vehicle matching degree described above; If the vehicle matching degree is greater than the preset vehicle matching threshold, then the vehicle matching degree is marked as the target matching degree, and the corresponding target monitoring point location and time are obtained; The final complete path is generated based on the corrected time window, the location of the target monitoring point, and the time.

4. A suspicious vehicle path tracking system based on spatiotemporal reverse retrieval, characterized in that, The system includes a memory and a processor. The memory contains a program for a suspicious vehicle path tracing method based on spatiotemporal reverse retrieval. When the program for the suspicious vehicle path tracing method based on spatiotemporal reverse retrieval is executed by the processor, it performs the following steps: Obtain road network structure information related to preset toll stations and vehicle feature database information of target suspicious vehicles; Extract path branches and match path priorities based on the road network structure information; Obtain the road segment scene data corresponding to each path branch, process it to obtain the vehicle speed probability density and the corrected time window; Acquire video data and process it in conjunction with the vehicle feature database information to obtain the vehicle matching degree; The final complete path is obtained based on the corrected time window and the vehicle matching degree. The acquisition of road network structure information related to preset toll stations and vehicle feature database information of target suspicious vehicles includes: If a suspicious vehicle is detected at a pre-set toll station, obtain the relevant road network structure information of the pre-set toll station and the vehicle feature database information of the suspicious vehicle. The road network structure information includes the destination location, destination time, road segment length, merging point information, and connection relationship information; The vehicle feature database information includes license plate number, vehicle model, color, vehicle wheel rims, and tire model; The step of extracting path branches and matching path priorities based on the road network structure information includes: The path branches are obtained by processing the endpoint location, endpoint time, road segment length, merging point information, and connection relationship information through a preset path prediction model. Obtain the historical driving data of the target suspicious vehicle, including branch selection records and the historical driving frequency corresponding to each path branch; Based on the historical driving data, the priority of each path branch is sorted to obtain the priority of each path branch; The process of acquiring road segment scene data corresponding to each path branch, and processing it to obtain vehicle speed probability density and corrected time windows includes: Obtain the road segment scene data corresponding to each path branch, including branch road segment length, historical vehicle speed distribution of the road segment, real-time traffic flow speed data, and real-time weather data; The vehicle speed probability density is obtained based on the historical vehicle speed distribution of the road segment. The time window is obtained by processing the branch segment length, vehicle speed probability density, and destination time. The time window is corrected based on the real-time traffic speed data and real-time weather data to obtain the corrected time window.

Images