Vehicle track identification method and device
By splitting and clustering the passing data of traffic violation vehicles, high-frequency trajectories are identified, and the problems of traffic police workload and low accuracy in the existing technology are solved, and the efficiency and accuracy of traffic safety management are improved.
Patent Information
- Application Number
- CN202310702138.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-13
- Publication Date
- 2024-12-13
AI Technical Summary
The existing technology lacks direct contact when identifying and analyzing the trajectory of traffic violation vehicles, resulting in traffic police workload, low accuracy and efficiency, making it difficult to effectively investigate and deal with hidden danger vehicles.
By obtaining the passing data of the target vehicle during the set period, splitting it into at least two track segments, and calculating the similarity between the track segments for clustering, the target trajectory in each category is determined to improve the target vehicle's size. Accuracy and efficiency of driving status analysis.
The accurate identification of the high-frequency trajectory of the target vehicle has been achieved, the efficiency of traffic police investigation and handling has been improved, the probability of traffic accidents has been reduced, and the traffic police have been assisted in the deployment of control.
Smart Images
Figure CN119132039A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of intelligent transportation technology, and in particular to a vehicle trajectory recognition method and device. Background Art
[0002] Order is the main factor affecting traffic safety. Usually, most traffic accidents are caused by traffic violations by traffic participants. There are many accidents caused by traffic violations, and there are great safety hazards. The vehicles that cause accidents are usually vehicles with safety hazards. The investigation and handling of vehicles with safety hazards is time-consuming and laborious, and usually requires manual investigation.
[0003] In the related technology, the locations where potential vehicles frequently appear are usually identified based on the passing vehicle data, but there is no direct connection between the locations. Therefore, if the traffic police want to investigate or deploy potential vehicles based on these location information alone, there are certain limitations, and the experience of the traffic police is also required. This results in a large workload for the traffic police, and the accuracy and efficiency of analyzing the state of the target vehicle are relatively low. Summary of the invention
[0004] In an exemplary embodiment of the present application, a vehicle trajectory recognition method and device are provided to determine the high-frequency trajectory of a target vehicle so as to improve the accuracy and efficiency of analyzing the driving state of the target vehicle.
[0005] According to a first aspect of an exemplary embodiment, a vehicle trajectory recognition method is provided, the method comprising at least the following steps:
[0006] Obtaining the passing data of the target vehicle within a set period; wherein the passing data includes the position and passing time of the point where the target vehicle passes;
[0007] Determine at least two trajectory segments included in the driving trajectory of the target vehicle within a set period according to the passing data of the target vehicle;
[0008] determining a similarity between at least two trajectory segments, and clustering the at least two trajectory segments according to the similarity;
[0009] According to the number of road segments included in the trajectory segment in each category, determine the target trajectory corresponding to the category;
[0010] Among the multiple target trajectories obtained from the multiple categories, at least one target trajectory is selected to analyze the driving state of the target vehicle.
[0011] According to a second aspect of an exemplary embodiment, there is provided a vehicle trajectory recognition device, comprising a data transmission unit and a processor;
[0012] The data transfer unit is configured to perform:
[0013] Obtaining the passing data of the target vehicle within a set period; wherein the passing data includes the position and passing time of the point where the target vehicle passes;
[0014] The processor is configured to execute:
[0015] Determine at least two trajectory segments included in the driving trajectory of the target vehicle within a set period according to the passing data of the target vehicle;
[0016] determining a similarity between at least two trajectory segments, and clustering the at least two trajectory segments according to the similarity;
[0017] According to the number of road segments included in the trajectory segment in each category, determine the target trajectory corresponding to the category;
[0018] At least one target trajectory is selected from a plurality of target trajectories obtained from a plurality of categories to analyze the driving state of a target vehicle. The terminal includes a processor and a transmission unit, wherein:
[0019] According to a third aspect of an exemplary embodiment, a vehicle trajectory recognition device is provided, the device comprising:
[0020] The data acquisition unit is used to: acquire the passing data of the target vehicle within a set period; wherein the passing data includes the position and passing time of the point where the target vehicle passes;
[0021] The processing unit is used to: determine at least two track segments included in the driving track of the target vehicle within a set period according to the passing data of the target vehicle;
[0022] The processing unit is further used to: determine the similarity between at least two trajectory segments, and cluster the at least two trajectory segments according to the similarity;
[0023] The processing unit is further used to: determine the target trajectory corresponding to the category according to the number of road segments included in the trajectory segment in each category;
[0024] The analysis unit is used to select at least one target trajectory from a plurality of target trajectories obtained from a plurality of categories to analyze the driving state of the target vehicle.
[0025] According to a fourth aspect of the exemplary implementation, a computer storage medium is provided, in which computer program instructions are stored. When the instructions are executed on a computer, the computer executes the method of the first aspect.
[0026] In the embodiment of the present application, the target vehicle's driving trajectory within the set period can be split into at least two trajectory segments according to the passing data of the target vehicle within the set period, and the analysis is performed with the trajectory segment of finer granularity as the dimension. The similarity between the at least two trajectory segments is then calculated, and then the at least two trajectory segments are clustered according to the similarity, so that the trajectory segments in each category can be guaranteed to be similar. In addition, according to the number of sections included in the trajectory segments in each category, the target trajectory that can represent the category is found in each category. The target trajectory corresponding to each category is the trajectory with a higher frequency of target vehicle occurrence. Therefore, among the multiple target trajectories obtained in multiple categories, a target trajectory can be selected to analyze the driving state of the target vehicle. Such a design determines the high-frequency trajectory of the target vehicle in order to improve the accuracy and efficiency of analyzing the driving state of the target vehicle. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0028] Figure 1 An application scenario diagram of a vehicle trajectory recognition method provided by an embodiment of the present application is exemplarily shown;
[0029] Figure 2 A flow chart of a vehicle trajectory recognition method provided by an embodiment of the present application is exemplarily shown;
[0030] Figure 3 A schematic diagram of dividing a driving trajectory into at least two trajectory segments provided by an embodiment of the present application is exemplarily shown;
[0031] Figure 4 A schematic diagram exemplarily shows a process of clustering at least two trajectory segments provided in an embodiment of the present application;
[0032] Figure 5 A schematic diagram of a clustering operation provided by an embodiment of the present application is exemplarily shown;
[0033] Figure 6 A schematic diagram exemplarily shows a process of determining a target trajectory provided by an embodiment of the present application;
[0034] Figure 7 A schematic diagram of a track segment search provided by an embodiment of the present application is exemplarily shown;
[0035] Figure 8A flowchart of a complete vehicle trajectory recognition method provided by an embodiment of the present application is exemplarily shown;
[0036] Fig. 9 The schematic diagram of the structure of a vehicle trajectory recognition device provided by an embodiment of the present application is exemplarily shown;
[0037] Fig.10 A schematic diagram of the structure of a vehicle trajectory recognition device provided in an embodiment of the present application is exemplified. DETAILED DESCRIPTION
[0038] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.
[0039] Order is the main factor affecting traffic safety. Usually, most traffic accidents are caused by traffic violations by traffic participants. There are many accidents caused by traffic violations, and there are great safety hazards. The vehicles that cause accidents are usually vehicles with safety hazards. The investigation and handling of vehicles with hidden dangers is time-consuming and laborious, and usually requires human investigation.
[0040] In the related technology, the locations where potential vehicles frequently appear are usually identified based on the passing vehicle data, but there is no direct connection between the locations. Therefore, if the traffic police want to investigate or deploy potential vehicles based on these location information alone, there are certain limitations, and the experience of the traffic police is also required. This results in a large workload for the traffic police, and the accuracy and efficiency of analyzing the state of the target vehicle are relatively low.
[0041] Optionally, vehicles with hidden dangers can be vehicles that have not been inspected within a set period, vehicles that have not been scrapped, overloaded vehicles, or overcrowded vehicles. Such vehicles are more likely to have accidents. Therefore, being able to accurately locate the frequently occurring trajectory segments of these vehicles can effectively assist traffic police in investigation or control, and can reduce the probability of traffic accidents to a certain extent. For vehicles without hidden dangers, their frequently occurring trajectory segments can also be used to analyze the driving behavior of the vehicle, such as which roads it often drives on.
[0042] For this reason, the embodiment of the present application provides a kind of track identification method of vehicle, in this method, can determine at least two track segments included in the driving track of target vehicle in the setting period according to the passing data of target vehicle in the setting period, then calculate the similarity between at least two track segments, and then cluster at least two track segments according to the similarity, so that the track segments in each category can be guaranteed to be similar. In addition, according to the number of sections included in the track segments in each category, determine the target track corresponding to the category. And the target track corresponding to each category is the track with more frequent occurrence of target vehicle, therefore, among the multiple target tracks obtained in multiple categories, a target track can be selected to analyze the driving state of target vehicle. Such design determines the high-frequency track of target vehicle, so as to analyze the driving state of target vehicle more accurately and efficiently.
[0043] After introducing the design ideas of the embodiments of the present application, the following briefly introduces the application scenarios to which the technical solutions of the embodiments of the present application can be applied. It should be noted that the application scenarios introduced below are only used to illustrate the embodiments of the present application and are not limited. In specific implementation, the technical solutions provided by the embodiments of the present application can be flexibly applied according to actual needs.
[0044] refer to Figure 1 , shows an application scenario diagram of a vehicle trajectory recognition method, in which three target vehicles are used as an example, abcdwf represents the driving trajectory of target vehicle 1 in the past week, mbcde represents the driving trajectory of target vehicle 2 in the past week, and qdch represents the driving trajectory of target vehicle 3 in the past week. In this example, trajectory segments cd and dc are high-frequency trajectory segments that each vehicle has passed through. The traffic police can be prompted to conduct investigations or control in this trajectory segment.
[0045] To further illustrate the technical solution provided by the embodiment of the present application, this is described in detail below in conjunction with the accompanying drawings and specific implementation methods. Although the embodiment of the present application provides the method operation steps shown in the following embodiments or drawings, more or fewer operation steps may be included in the method based on routine or no creative labor. In the steps where there is no necessary causal relationship logically, the execution order of these steps is not limited to the execution order provided in the embodiment of the present application.
[0046] Combine the following Figure 1 The application scenario shown is shown in Figure 2 A flow chart of a vehicle trajectory recognition method is shown, and the method can be executed by a trajectory device, and the trajectory recognition device can be an electronic device or a server with data processing function. Next, the technical solution provided by the embodiment of the present application is described.
[0047] S201: The trajectory recognition device obtains the passing data of the target vehicle within a set period.
[0048] The vehicle passing data includes the location and passing time of the point where the target vehicle passes.
[0049] S202: The trajectory recognition device determines at least two trajectory segments included in the driving trajectory of the target vehicle within a set period according to the passing data of the target vehicle.
[0050] S203: The trajectory recognition device determines the similarity between at least two trajectory segments, and clusters the at least two trajectory segments according to the similarity.
[0051] S204: The trajectory recognition device determines a target trajectory corresponding to a category according to the number of road segments included in the trajectory segments in each category.
[0052] S205: The trajectory recognition device selects at least one target trajectory from a plurality of target trajectories obtained from a plurality of categories to analyze the driving state of the target vehicle.
[0053] In the embodiment of the present application, the target vehicle's driving trajectory within the set period can be split into at least two trajectory segments according to the passing data of the target vehicle within the set period, and the analysis is performed with the trajectory segment of finer granularity as the dimension. The similarity between the at least two trajectory segments is then calculated, and then the at least two trajectory segments are clustered according to the similarity, so that the trajectory segments in each category can be guaranteed to be similar. In addition, according to the number of sections included in the trajectory segment in each category, the target trajectory that can represent the category is found in each category. The target trajectory corresponding to each category is the trajectory with a higher frequency of target vehicle occurrence. Therefore, among the multiple target trajectories obtained in multiple categories, a target trajectory can be selected to analyze the driving state of the target vehicle. Such a design determines the high-frequency trajectory of the target vehicle so as to analyze the driving state of the target vehicle more accurately and efficiently.
[0054] Regarding S201, capture devices, such as bayonet cameras, are usually set at intervals on the road. The capture devices upload the captured images or videos to the track recognition device so that the track recognition device can process the data. In the embodiment of the present application, in order to improve accuracy, usually only the data within a period of time in the past is analyzed. For example, the set period is 1 week, that is, one week before the current moment. The track recognition device obtains the passing data of the target vehicle within the set time period, wherein the passing data includes the position and passing time of the point where the target vehicle passes. In the actual application process, a capture device is deployed at each point, and the position of the point where the target vehicle passes is the position of the capture device that photographed the target vehicle, and the passing time is the capture time. Optionally, the target vehicle can be a normal driving vehicle or a vehicle with hidden dangers. In the embodiment of the present application, in order to assist the traffic police in investigation and handling, the target vehicle can be a vehicle with hidden dangers. In a specific example, there can be multiple target vehicles.
[0055] In S202, for each target vehicle, the trajectory recognition device can determine at least two trajectory segments included in the driving trajectory of the target vehicle within a set period based on the passing data of the target vehicle.
[0056] In this process, the process of obtaining the track segment according to the vehicle passing data can be realized through step A1:
[0057] A1: If, in the target vehicle's passing data, there are two adjacent points whose passing time is greater than a time threshold, the target vehicle's driving trajectory within a set period is divided into at least two trajectory segments according to the positions of the two adjacent points.
[0058] Normally, in urban traffic, there are relatively dense capture devices, that is, the interval between the passing time of two adjacent points will not be particularly long. When the passing time of two adjacent points in the target vehicle's passing data is greater than the time threshold, it can be determined that the target vehicle has stopped. At this time, the larger passing time of the two adjacent points can be used as the split point, and the driving trajectory of the target vehicle within the set period can be divided into at least two trajectory segments.
[0059] Figure 3 A schematic diagram of dividing a driving trajectory into at least two trajectory segments is provided in an embodiment of the present application, wherein the starting point of the driving trajectory is M, the ending point is N, and the point M is 1 The elapsed time is t 1 , point M 2 The elapsed time is t 2 , and t 1 With t 2 If the time difference is greater than a time threshold (e.g., 30 minutes), the driving trajectory MN is divided into trajectory segments MM. 1 With trajectory segment M2 N. For any track segment, for example, it includes p points, the vehicle passing data of the track segment is recorded as { <t 1 ,s 1 >, <t 2 ,s 2 >, …, <t p ,s p >}.
[0060] In S203, for each target vehicle, the method of step S202 is applied to obtain at least two track segments included in the driving track of the target vehicle within a set period. When there are multiple target vehicles, multiple track segments can be obtained.
[0061] In order to determine the high-frequency trajectory segments that each target vehicle has passed through multiple times, the similarity between at least two trajectory segments is calculated with the trajectory segment as the granularity, and then each trajectory segment is clustered according to the similarity. Optionally, the higher the similarity, the higher the overlap between the two trajectory segments, and at least two trajectories with high similarity can be classified into the same category, indicating that the overlap between the trajectory segments in this category is very high.
[0062] Optionally, you can Figure 4 Steps S203-1 and S203-2 in the embodiment implement the process of clustering at least two trajectory segments according to similarity:
[0063] S203-1: The trajectory recognition device performs the following clustering operation on any trajectory segment to determine the trajectory segments that belong to the same category as the trajectory segment: determine the similarity between the trajectory segment and each of the remaining trajectory segments, and if the obtained similarity meets the similarity condition, determine that the trajectory segment corresponding to the similarity that meets the condition and the trajectory segment are both first-category trajectory segments.
[0064] Here, it is assumed that there are 10 trajectory segments, which are numbered 1-10 respectively, and trajectory segment 1 is used as an example for explanation. The similarity between trajectory segment 1 and the remaining trajectory segments 2-10 is calculated, and 9 similarities are obtained. When the 9 similarities meet the similarity condition, the trajectory segments corresponding to the similarities that meet the condition and trajectory segment 1 can be classified into the first category of trajectory segments. For example, the first category of trajectory segments includes trajectory segments 1, 4, 8, and 10.
[0065] Optionally, the similarity condition may be that the first similarity and the second similarity satisfy the first condition (e.g., the first similarity is greater than 80% and the third similarity is greater than 70%), or that the second similarity and the third similarity satisfy the second condition (e.g., the second similarity is greater than 80% and the third similarity is greater than 70%). Exemplarily, the similarity between two trajectories may be determined through steps B1-B2:
[0066] B1: The trajectory recognition device obtains a first number of points included in the first trajectory segment, a second number of points included in the second trajectory segment, and a third number of points where the first trajectory segment and the second trajectory segment overlap.
[0067] The first track segment and the second track segment are any two different track segments. 1 The number of points included is g 1 , the second trajectory segment G 2 The number of points included is g 2 , the third track segment G 3 The number of points included is g 3 .
[0068] B2: The trajectory recognition device calculates a first similarity according to the first point number and the third point number, calculates a second similarity according to the second point number and the third point number, and calculates a third similarity according to the first point number, the second point number and the third point number.
[0069] Optionally, the first similarity S 1 =g 3 / g 1 , the second similarity S 2 =g 3 / g 2 , the third similarity S 1 =2g 3 / (g 1 +g 2 ).
[0070] In the embodiment of the present application, the first similarity is based on the first track segment to describe the similarity between the first track segment and the second track segment; the second similarity is based on the second track segment to describe the similarity between the first track segment and the second track segment; the third similarity is based on the first track segment and the second track segment to describe the similarity between the first track segment and the second track segment. Compared with comparing only one similarity, the accuracy is improved.
[0071] S203 - 2 : The trajectory recognition device continues to perform clustering operations on all trajectory segments except for the trajectory segments included in the first category of trajectory segments, until each trajectory segment belongs to one category.
[0072] Continuing to refer to the above embodiment, clustering operation is continued for all the trajectories except the remaining trajectory segments included in the first category of trajectory segments, such as trajectory segments 2, 3, 5, 6, and 7, until each trajectory segment belongs to the same category.
[0073] In addition, the clustering operation can also be implemented in the following way: after obtaining the first category of trajectory segments, loop in sequence, and skip if the current trajectory segment already appears in the existing category, that is, ensure that each trajectory segment has a unique category. Figure 5 A schematic diagram of a clustering operation provided in an embodiment of the present application, wherein a total of n trajectory segments are included, numbered 1-n respectively.
[0074] In S204, since the track segments in each category are very similar, a target track that can represent the category is found among these similar track segments. Optionally, the track recognition device determines the target track corresponding to the category according to the number of road segments included in the track segments in each category.
[0075] Among them, it can be Figure 6 Steps S204-1 to S204-3 in the process of determining the target trajectory are implemented as follows:
[0076] S204-1: The trajectory recognition device determines the starting search point of each trajectory segment included in the category.
[0077] Take the first type of track segment as an example, including track segments 1, 4, 8 and 10. In the process of determining the target track, it is necessary to first locate a starting search point, and then start searching from the point.
[0078] Optionally, the specific process of step S204-1 can be implemented through steps C1-C2.
[0079] C1: The trajectory recognition device obtains the number of occurrences of each road segment included in each trajectory segment included in the category.
[0080] In a trajectory segment, two adjacent points constitute a road segment. Therefore, the trajectory recognition device can determine the road segments included in each trajectory segment, and further determine the road segments included in each trajectory and the number of times each of them appears.
[0081] For example, if the point included in track segment 1 is abc, then the road segments included in track segment 1 are ab and bc; if the point included in track segment 2 is abefg, then the road segments included in track segment 2 are ab, be, ef and fg... "-" indicates the driving direction, that is, road segments ab and ba can be regarded as two different road segments.
[0082] Table 1 is a schematic table of the number of times of each road section provided in the embodiment of the present application
[0083] The starting point number of the road segment The end point number of the road segment Number of times the road segment appears a b 30 a c 25 b d 28 b e 32
[0084] In another specific example, Figure 7 A search schematic diagram provided in the embodiment of the present application, refer to Figure 7 In the first type of trajectory segments, the number of segment ab is 20 times, the number of segment bc is 25 times, the number of segment cd is 26 times, the number of segment de is 20 times, the number of segment ef is 18 times... In this step, the number of times each segment appears is obtained.
[0085] C2: The trajectory recognition device uses the end point of the road segment that appears most times as the starting point for searching.
[0086] Still refer to Figure 7 , the road segment with the most appearances is bc, therefore, the trajectory recognition device uses the end point c of the road segment bc with the most appearances as the starting search point.
[0087] S204-2: The trajectory recognition device starts from the starting search point, determines at least one road section starting from the starting search point, and calculates the number of times the at least one road section appears.
[0088] The track recognition device starts from point c, determines at least one road section starting from point c, for example, including road section cd and road section ce, and calculates the number of times road section cd and road section ce appear respectively.
[0089] S204-3: The trajectory recognition device selects the end point of the road segment with the highest number of appearances as a new starting search point, and repeats the search operation until the number of appearances of the new road segment is less than the set threshold, and connects the searched road segments in chronological order to obtain the target trajectory corresponding to the category.
[0090] Optionally, if the number of times segment cd appears is 26 and the number of times segment ce appears is 15, then the end point d of segment cd is selected as the new starting point for the search, and the search operation is repeated until the number of times the new segment appears is less than the set threshold (e.g. 16), and the searched segments are connected in chronological order to obtain the target trajectory corresponding to the first type of trajectory segment. Figure 7 In the example, the target trajectory is bcdef.
[0091] The above process is the process of determining the target trajectory of a category of trajectory segments. In actual application, if there are multiple categories, the target trajectories corresponding to the multiple categories can be obtained. If the number of trajectory segments included in the partial cluster is small, the data of this category can be discarded and not processed.
[0092] In S205, the trajectory recognition device selects at least one target trajectory from a plurality of target trajectories obtained from a plurality of categories to analyze the driving state of the target vehicle.
[0093] Optionally, at least one target trajectory can be selected from multiple target trajectories according to the number of points included in each target trajectory or other indicators. The selected at least one target trajectory is used to analyze the state of the target vehicle. The analysis here can be, for example, analyzing the time and frequency of different target vehicles appearing on the selected at least one target trajectory. With such a design, when the target vehicle is a hidden danger vehicle, the number of monitoring devices or the operating frequency of the monitoring devices on the road corresponding to the selected at least one target trajectory can be adjusted, which can effectively guide the traffic police to control these roads, etc., so as to reduce the probability of traffic accidents, and can also improve the efficiency of illegal capture.
[0094] It should be noted that the data examples in the embodiments of the present application are only for illustration and do not constitute specific limitations.
[0095] In the embodiment of the present application, the passing data of vehicles with hidden dangers are processed, and the various track segments included in the track are clustered. The track segments in each category are similar, and then in each category, the target track under this category is found as the high-frequency track. Such a design can accurately identify high-frequency tracks, improve the efficiency of traffic police investigation and provide strong support for scheduling and deployment.
[0096] In order to make the technical solution of the present application more perfect, the following is a complete flowchart to illustrate the vehicle trajectory recognition method in the embodiment of the present application, with reference to Figure 8 .
[0097] S801: The trajectory recognition device obtains the passing data of the target vehicle within a set period.
[0098] The implementation method of this step can refer to step S201 and will not be repeated here.
[0099] S802: If the passing time of two adjacent points in the target vehicle's passing data is greater than the time threshold, the trajectory recognition device divides the target vehicle's driving trajectory within a set period into at least two trajectory segments according to the positions of the two adjacent points.
[0100] The implementation method of this step can refer to step A1, which will not be repeated here.
[0101] S803: The trajectory recognition device performs the following clustering operation on any trajectory segment to determine the trajectory segments that belong to the same category as the trajectory segment: determine if the similarity between the trajectory segment and each of the remaining trajectory segments satisfies a similarity condition, then determine that the trajectory segments corresponding to the similarities that meet the condition and the trajectory segments are both first category trajectory segments.
[0102] The implementation method of this step can refer to step S203-1, which will not be repeated here.
[0103] S804: The trajectory recognition device continues to perform clustering operations on all trajectory segments except for the trajectory segments included in the first category of trajectory segments, until each trajectory segment belongs to one category.
[0104] The implementation method of this step can refer to step S203-2, which will not be repeated here.
[0105] S805: The trajectory recognition device obtains the number of occurrences of each road segment included in each trajectory segment included in each category.
[0106] The implementation method of this step can refer to step D11, which will not be repeated here.
[0107] S806: The trajectory recognition device uses the end point of the road section that appears most times as the starting search point.
[0108] The implementation method of this step can refer to step D12, which will not be repeated here.
[0109] S807: The trajectory recognition device starts from the starting search point, determines at least one road section starting from the starting search point, and calculates the number of times the at least one road section appears.
[0110] The implementation method of this step can refer to step S204-2, which will not be repeated here.
[0111] S808: The trajectory recognition device selects the end point of the road segment with the highest number of appearances as a new starting search point, and repeats the search operation until the number of appearances of the new road segment is less than the set threshold, and connects the searched road segments in chronological order to obtain the target trajectory corresponding to the category.
[0112] The implementation method of this step can be found in step S204-3, which will not be described in detail here.
[0113] S809: The trajectory recognition device selects at least one target trajectory from a plurality of target trajectories obtained from a plurality of categories to analyze the driving state of the target vehicle.
[0114] The implementation method of this step can refer to step S204, which will not be repeated here.
[0115] S810: The trajectory recognition device adjusts the number of monitoring devices or the operating frequency of the monitoring devices on the road corresponding to at least one target trajectory.
[0116] The implementation method of this step can refer to step S205, which will not be repeated here.
[0117] In the embodiment of the present application, the trajectory recognition device analyzes and clusters the vehicle passing data and classifies similar trajectory segments into one category. Since each category includes multiple trajectory segments, the target trajectory segment that best represents the category is found as the target trajectory, that is, the high-frequency trajectory of the target vehicle. On the one hand, the driving status of the target vehicle can be analyzed based on the high-frequency trajectory. On the other hand, the efficiency of capturing illegal activities of the target vehicle can be improved by adjusting the number of monitoring devices on the road corresponding to the target trajectory or the operating frequency of the monitoring devices.
[0118] like Fig. 9 As shown, based on the same inventive concept, an embodiment of the present application provides a trajectory recognition device, including a data acquisition unit 91 , a processing unit 92 and an analysis unit 93 .
[0119] The data acquisition unit 91 is used to: acquire the passing data of the target vehicle within a set period; wherein the passing data includes the position and passing time of the point where the target vehicle passes;
[0120] The processing unit 92 is used to: determine at least two track segments included in the driving track of the target vehicle within a set period according to the passing data of the target vehicle;
[0121] The processing unit 92 is further configured to: determine a similarity between at least two trajectory segments, and cluster the at least two trajectory segments according to the similarity;
[0122] The processing unit 92 is further used to: determine the target trajectory corresponding to the category according to the number of road segments included in the trajectory segment in each category;
[0123] The analysis unit 93 is used to select at least one target trajectory from among the multiple target trajectories obtained from the multiple categories to analyze the driving state of the target vehicle.
[0124] In an optional implementation, the processing unit 92 is specifically configured to:
[0125] If, in the target vehicle's passing data, there are two adjacent points whose passing time is greater than a time threshold, the target vehicle's driving trajectory within a set period is divided into at least two trajectory segments according to the positions of the two adjacent points.
[0126] In an optional implementation, the processing unit 92 is specifically configured to:
[0127] For any trajectory segment, the following clustering operation is performed to determine the trajectory segments that belong to the same category as the trajectory segment: determine that the similarity between the trajectory segment and each of the remaining trajectory segments meets the similarity condition, then determine that the trajectory segment corresponding to the similarity that meets the condition and the trajectory segment are both first-category trajectory segments;
[0128] Among all the trajectory segments, except for the trajectory segments included in the first category, the clustering operation is continued until each trajectory segment belongs to one category.
[0129] In an optional implementation, the similarity condition is that if the first similarity and the second similarity satisfy the first condition, or the second similarity and the third similarity satisfy the second condition;
[0130] The processing unit 92 is specifically used for:
[0131] Obtaining a first number of points included in the first track segment, a second number of points included in the second track segment, and a third number of points where the first track segment and the second track segment overlap; wherein the first track segment and the second track segment are any two different track segments;
[0132] A first similarity is calculated based on the first point number and the third point number, a second similarity is calculated based on the second point number and the third point number, and a third similarity is calculated based on the first point number, the second point number and the third point number.
[0133] In an optional implementation, the processing unit 92 is specifically configured to:
[0134] Determine the starting search point of each track segment included in the category;
[0135] Starting from the starting search point, determining at least one road section starting from the starting search point, and calculating the number of times the at least one road section appears;
[0136] The end point of the road segment with the highest number of appearances is selected as the new starting point for searching. The search operation is repeated until the number of appearances of the new road segment is less than the set threshold. The found road segments are connected in chronological order to obtain the target trajectory corresponding to the category.
[0137] In an optional implementation, the processing unit 92 is specifically configured to:
[0138] Obtain the number of occurrences of each road segment included in each trajectory segment included in the category;
[0139] The end point of the road segment that appears most times is used as the starting point for searching.
[0140] In an optional implementation, the target vehicles include vehicles that are overdue for annual inspection, vehicles that are overdue for scrapping, overloaded vehicles, or overcrowded vehicles within a set period.
[0141] In an optional implementation, the system further includes an adjustment unit for adjusting, after obtaining a plurality of target trajectories, the number of monitoring devices on a road corresponding to at least one target trajectory or the operating frequency of the monitoring devices.
[0142] Since the device is the device in the method in the embodiment of the present application, and the principle of solving the problem by the device is similar to that of the method, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.
[0143] like Fig.10 As shown, based on the same inventive concept, an embodiment of the present application provides a trajectory recognition device, which includes: a processor 1001 and a data transmission unit 1002.
[0144] The data transmission unit 1002 is configured to perform:
[0145] Obtaining the passing data of the target vehicle within a set period; wherein the passing data includes the position and passing time of the point where the target vehicle passes;
[0146] The processor 1001 is configured to execute:
[0147] Determine at least two trajectory segments included in the driving trajectory of the target vehicle within a set period according to the passing data of the target vehicle;
[0148] determining a similarity between at least two trajectory segments, and clustering the at least two trajectory segments according to the similarity;
[0149] According to the number of road segments included in the trajectory segment in each category, determine the target trajectory corresponding to the category;
[0150] Among the multiple target trajectories obtained from the multiple categories, at least one target trajectory is selected to analyze the driving state of the target vehicle.
[0151] In an optional implementation, the processor 1001 is specifically configured to execute:
[0152] If in the target vehicle's passing data, there are two adjacent points whose passing time is greater than the time threshold, the target vehicle's driving trajectory within the set period is divided into at least two trajectory segments with the first time as the dividing point; wherein the first time is the larger time of the passing time of the two adjacent points.
[0153] In an optional implementation, the processor 1001 is specifically configured to execute:
[0154] For any trajectory segment, the following clustering operation is performed to determine the trajectory segments that belong to the same category as the trajectory segment: determine that the similarity between the trajectory segment and each of the remaining trajectory segments meets the similarity condition, then determine that the trajectory segment corresponding to the similarity that meets the condition and the trajectory segment are both first-category trajectory segments;
[0155] Among all the trajectory segments, except for the trajectory segments included in the first category, the clustering operation is continued until each trajectory segment belongs to one category.
[0156] In an optional implementation, the similarity condition is that if the first similarity and the second similarity satisfy the first condition, or the second similarity and the third similarity satisfy the second condition;
[0157] The processor 1001 is specifically configured to execute:
[0158] Obtaining a first number of points included in the first track segment, a second number of points included in the second track segment, and a third number of points where the first track segment and the second track segment overlap; wherein the first track segment and the second track segment are any two different track segments;
[0159] A first similarity is calculated based on the first point number and the third point number, a second similarity is calculated based on the second point number and the third point number, and a third similarity is calculated based on the first point number, the second point number and the third point number.
[0160] In an optional implementation, the processor 1001 is specifically configured to execute:
[0161] Determine the starting search point of each track segment included in the category;
[0162] Starting from the starting search point, determining at least one road section starting from the starting search point, and calculating the number of times the at least one road section appears;
[0163] The end point of the road segment with the highest number of appearances is selected as the new starting point for searching. The search operation is repeated until the number of appearances of the new road segment is less than the set threshold. The found road segments are connected in chronological order to obtain the target trajectory corresponding to the category.
[0164] In an optional implementation, the processor 1001 is specifically configured to execute:
[0165] Obtain the number of occurrences of each road segment included in each trajectory segment included in the category;
[0166] The end point of the road segment that appears most times is used as the starting point for searching.
[0167] In an optional implementation, the target vehicles include vehicles that are overdue for annual inspection, vehicles that are overdue for scrapping, overloaded vehicles, or overcrowded vehicles within a set period.
[0168] In an optional implementation, after obtaining multiple target trajectories, the processor 1001 is further configured to execute:
[0169] The number of monitoring devices or the operating frequency of the monitoring devices on the road corresponding to at least one target trajectory is adjusted.
[0170] The embodiment of the present application further provides a computer storage medium, in which computer program instructions are stored. When the instructions are executed on a computer, the computer executes the steps of the above-mentioned trajectory recognition method.
[0171] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.
[0172] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0173] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0174] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0175] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.
Claims
1. A vehicle trajectory recognition method, characterized in that: include: Obtaining the passing data of the target vehicle within a set period; wherein the passing data includes the position and passing time of the point where the target vehicle passes; Determine, according to the passing data of the target vehicle, at least two track segments included in the driving track of the target vehicle within the set period; Determining a similarity between the at least two trajectory segments, and clustering the at least two trajectory segments according to the similarity; Determine the target trajectory corresponding to each category according to the number of road segments included in the trajectory segment in the category; Among the multiple target trajectories obtained from the multiple categories, at least one target trajectory is selected to analyze the driving state of the target vehicle.
2. The method according to claim 1, characterized in that The step of determining, based on the passing data of the target vehicle, at least two trajectory segments included in the driving trajectory of the target vehicle within the set period comprises: If, in the passing data of the target vehicle, there are two adjacent points whose passing time is greater than a time threshold, the driving trajectory of the target vehicle within the set period is divided into at least two trajectory segments according to the positions of the two adjacent points.
3. The method according to claim 1, characterized in that The clustering of the at least two trajectory segments according to the similarity: For any trajectory segment, the following clustering operation is performed to determine the trajectory segment that belongs to the same category as the trajectory segment: determine that the similarity between the trajectory segment and each of the remaining trajectory segments satisfies the similarity condition, and then determine that the trajectory segment corresponding to the similarity that satisfies the condition and the trajectory segment are both first-category trajectory segments; In all the trajectory segments, except for the trajectory segments included in the first category of trajectory segments, the clustering operation is continued to be performed until each trajectory segment belongs to one category.
4. The method according to claim 3, characterized in that The similarity condition is that if the first similarity and the second similarity satisfy a first condition, or the second similarity and the third similarity satisfy a second condition; Determining the similarity between the at least two trajectory segments: Obtain a first number of points included in the first track segment, a second number of points included in the second track segment, and a third number of points overlapped by the first track segment and the second track segment; wherein the first track segment and the second track segment are any two different track segments; A first similarity is calculated based on the first point number and the third point number, a second similarity is calculated based on the second point number and the third point number, and a third similarity is calculated based on the first point number, the second point number and the third point number.
5. The method according to claim 1, characterized in that The step of determining the target trajectory corresponding to each category according to the number of road segments included in the trajectory segments in each category includes: Determining the starting search point of each track segment included in the category; Starting from the starting search point, determining at least one road section starting from the starting search point, and calculating the number of times the at least one road section appears; The end point of the road segment with the highest number of appearances is selected as the new starting search point, and the search operation is repeated until the number of appearances of the new road segment is less than the set threshold, and the searched road segments are connected in chronological order to obtain the target trajectory corresponding to the category.
6. The method according to claim 5, characterized in that The step of determining the starting search point of each track segment included in the category includes: Obtaining the number of occurrences of each road segment included in each trajectory segment included in the category; The end point of the road segment that appears most times is used as the starting point for searching.
7. The method according to claim 1, characterized in that The target vehicles include vehicles that are overdue for annual inspection, vehicles that are overdue for scrapping, overloaded vehicles or overcrowded vehicles within the set period.
8. The method according to any one of claims 1 to 7, characterized in that: After obtaining a plurality of target trajectories, the method further includes: The number of monitoring devices or the operating frequency of the monitoring devices on the road corresponding to at least one target trajectory is adjusted.
9. A vehicle trajectory recognition device, characterized in that: including a data transmission unit and a processor; The data transmission unit is configured to perform: Obtaining the passing data of the target vehicle within a set period; wherein the passing data includes the position and passing time of the point where the target vehicle passes; The processor is configured to perform: Determine, according to the passing data of the target vehicle, at least two track segments included in the driving track of the target vehicle within the set period; Determining a similarity between the at least two trajectory segments, and clustering the at least two trajectory segments according to the similarity; Determine the target trajectory corresponding to each category according to the number of road segments included in the trajectory segment in the category; Among the multiple target trajectories obtained from the multiple categories, at least one target trajectory is selected to analyze the driving state of the target vehicle.
10. The device according to claim 9, characterized in that The processor is specifically configured to execute: If in the passing data of the target vehicle, there are two adjacent points whose passing time is greater than the time threshold, the driving trajectory of the target vehicle within the set period is divided into at least two trajectory segments with the first time as the dividing point; wherein the first time is the larger time of the passing time of the two adjacent points.