Spatial-temporal information recommendation method and device, equipment and storage medium

Abstract

The invention provides a spatio-temporal information recommendation method and device, equipment and a storage medium, and relates to the technical field of traffic safety. The method comprises the following steps: analyzing a video in a preset area and historical vehicle passing data acquired by bayonet equipment to obtain multiple pieces of spatio-temporal data, and determining spatio-temporal information corresponding to each piece of spatio-temporal data; clustering the spatio-temporal information according to the similarity among the spatio-temporal information, and determining the risk weight of each category based on the quantity of the spatio-temporal information in each category; and determining to-be-recommended spatio-temporal information from the spatio-temporal information based on the risk weight of each category. By means of the method, the accuracy of the predicted time and point location of the target vehicle is guaranteed, and meanwhile the troubleshooting efficiency of the target vehicle can be improved.

Description

Technical Field

[0001] The present application relates to the field of traffic safety technology, and in particular to a method, device, equipment and storage medium for recommending spatiotemporal information. Background Art

[0002] According to published data, most traffic accidents are caused by violations of regulations. Currently, violations of regulations are mainly investigated on-site and off-site. Off-site investigations are carried out through video and checkpoint equipment after violations such as running a red light occur. In addition, it is impossible to accurately analyze vehicles that have not been inspected or scrapped within the time limit. On-site investigations can handle such vehicles, but the current spatial and temporal deployment of personnel mainly relies on manual experience and has low investigation efficiency. Accurately predicting the spatial and temporal patterns of travel of vehicles that violate regulations will help improve the efficiency of investigating vehicles that violate regulations and reduce the occurrence of traffic accidents. Summary of the invention

[0003] The embodiments of the present application provide a method, apparatus, device and storage medium for recommending spatiotemporal information, which can predict the future spatiotemporal information of a target vehicle.

[0004] In a first aspect, an embodiment of the present application provides a method for recommending spatiotemporal information, the method comprising:

[0005] Parse the video and historical vehicle passing data acquired by the checkpoint device in the preset area to obtain multiple spatiotemporal data, and determine the spatiotemporal information corresponding to each spatiotemporal data, wherein each spatiotemporal data includes an image of a vehicle that does not comply with the preset driving rules and / or an image of a vehicle with a target identification, and the spatiotemporal information includes the time and point of the corresponding spatiotemporal data, the time of the spatiotemporal data is the time of the historical vehicle passing data acquired by the video and checkpoint device, and the point of the spatiotemporal data is the position of the video and checkpoint device that collected the spatiotemporal data;

[0006] Clustering the spatiotemporal information according to the similarity between the spatiotemporal information, and determining the risk weight of each category based on the amount of spatiotemporal information in each category;

[0007] Based on the risk weight of each category, the spatiotemporal information to be recommended is determined from the spatiotemporal information.

[0008] In the above embodiment, the historical vehicle passing data of vehicles that do not meet the preset driving rules and vehicles with target identification are comprehensively analyzed, and the analysis dimension is more refined; the spatiotemporal information is clustered, and based on the risk weight of each category, the spatiotemporal information to be recommended is determined from the clustered spatiotemporal information. Through the above embodiment, the time and location where vehicles that do not meet the preset driving rules and vehicles with target identification will appear can be accurately predicted.

[0009] In a possible implementation manner, before parsing the video in the preset area and the historical vehicle passing data acquired by the checkpoint device to obtain the plurality of spatiotemporal data, the method further includes:

[0010] Determine that the current computing power of the electronic device is less than a preset computing power value.

[0011] In the above embodiment, before parsing the historical vehicle passing data, by determining that the computing power of the current electronic device is less than the preset computing power value, a method for selecting the recommended spatiotemporal information that meets the current computing power can be determined, thereby avoiding the problem that the current computing power is insufficient to support the execution of the recommended method due to improper selection of the recommended method.

[0012] In a possible implementation, the method further includes:

[0013] If the computing power of the current electronic device is not less than the preset computing power value, the historical vehicle passing data acquired by the video and the checkpoint device in the preset area is analyzed to obtain the historical travel information of the target vehicle, wherein the target vehicle includes a vehicle that does not comply with the preset driving rules and a vehicle with a target identification;

[0014] If the spatiotemporal data of any target vehicle is obtained through the video and the vehicle passing data currently acquired by the checkpoint device in the preset area, the current spatiotemporal information of the any target vehicle is determined based on the spatiotemporal data of the any target vehicle;

[0015] Based on the historical travel information of the target vehicle and the current spatiotemporal information, the future spatiotemporal information of any one of the target vehicles is predicted.

[0016] In the above embodiment, if the computing power of the current electronic device is not less than the preset computing power value, the future spatiotemporal information of any target vehicle is predicted using the historical vehicle passing data and the current spatiotemporal information of the target vehicle. The embodiment of the present application adopts a real-time calculation and prediction method to make the predicted future spatiotemporal information more accurate.

[0017] In a possible implementation, the step of analyzing the video in the preset area and the historical vehicle passing data acquired by the checkpoint device to obtain the historical travel information of the target vehicle includes:

[0018] For any target vehicle, filter out the historical vehicle passing data of the target vehicle from the historical vehicle passing data;

[0019] The historical vehicle passing data of any one of the target vehicles is analyzed to obtain the historical travel information of any one of the target vehicles.

[0020] In the above embodiment, each target vehicle corresponds to a historical travel information, so that when predicting future time information, the target vehicle can be predicted based on its own historical travel information, thereby improving the accuracy of the prediction.

[0021] In a possible implementation manner, parsing the historical vehicle passing data of any one of the target vehicles to obtain the historical travel information of any one of the target vehicles includes:

[0022] Determine at least one driving route of any one of the target vehicles according to the historical vehicle passing data of any one of the target vehicles, wherein each first driving route includes the driving direction of any one of the target vehicles, the points passed by any one of the target vehicles, and the time of arrival at each point;

[0023] For any driving route, according to the driving direction of the any driving route, the number of times corresponding to each adjacent point on the any driving route is counted, wherein the number of times corresponding to the adjacent point is the number of times the any target vehicle reaches another point from one of the adjacent points according to the driving direction;

[0024] For any adjacent point, based on the number of times corresponding to the any adjacent point, determine the transfer probability of the any target vehicle from one of the adjacent points to another point according to the driving direction;

[0025] The historical travel information of any one of the target vehicles is determined based on the transition probability between adjacent points on the at least one route.

[0026] In the above embodiment, historical vehicle passing data is used to obtain historical travel information including the target vehicle's travel direction on the route, the points it passes through, the time it arrives at each point, and the transition probability between adjacent points, so that the future spatiotemporal information of the target vehicle can be predicted through the historical travel information.

[0027] In a possible implementation manner, predicting the future spatiotemporal information of any one of the target vehicles based on the historical travel information of the target vehicle and the current spatiotemporal information includes:

[0028] For any target vehicle, query the historical travel information of any target vehicle for target spatiotemporal information that matches the current spatiotemporal information; if no target information is found, query the historical travel information of other target vehicles other than the any target vehicle for target spatiotemporal information that matches the current spatiotemporal information;

[0029] The future spatiotemporal information of any target vehicle is predicted based on the transfer probability of the point corresponding to the queried target spatiotemporal information to the next point and the time included in the target spatiotemporal information.

[0030] In the above embodiment, for any target vehicle, if its own historical travel information does not meet the prediction requirements, the historical travel information of other target vehicles is used for prediction instead of directly ending the process. The analysis of the embodiment of the present application is more comprehensive and can ensure the accuracy of the prediction.

[0031] In a possible implementation, clustering the various spatiotemporal information according to the similarity between the various spatiotemporal information includes:

[0032] Each piece of space-time information is converted into a corresponding space-time vector, and the space-time information corresponding to each space-time vector is clustered according to the distance between the space-time vectors.

[0033] In the above embodiment, the time-space vectors are clustered instead of just single-dimensional clustering, so that in the recommendation step, both the time and the point can be recommended simultaneously, so that when checking the target vehicle, the specific point can be reached at a specific time for checking, thereby improving the efficiency of checking the target vehicle.

[0034] In a second aspect, an embodiment of the present application provides a device for recommending spatiotemporal information, the device comprising:

[0035] A parsing module is used to parse the video and historical vehicle passing data acquired by the checkpoint device in the preset area to obtain multiple spatiotemporal data, and determine the spatiotemporal information corresponding to each spatiotemporal data, wherein each spatiotemporal data includes an image of a vehicle that does not comply with the preset driving rules and / or an image of a vehicle with a target identification, and the spatiotemporal information includes the time and point of the corresponding spatiotemporal data, the time of the spatiotemporal data is the time when the video and checkpoint device collect the spatiotemporal data, and the point of the spatiotemporal data is the position of the video and checkpoint device that collect the spatiotemporal data;

[0036] A clustering module, for clustering the spatiotemporal information according to the similarity between the spatiotemporal information, and determining the risk weight of each category based on the amount of spatiotemporal information in each category;

[0037] A recommendation module is used to determine the spatiotemporal information to be recommended from the spatiotemporal information based on the risk weight of each category.

[0038] In a third aspect, an embodiment of the present application provides an electronic device, the device comprising:

[0039] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method of the first aspect as described above.

[0040] In a fourth aspect, an embodiment of the present application provides a computer storage medium, wherein the computer storage medium stores a computer program, and the computer program is used to enable a computer to execute the method of the first aspect as described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A schematic diagram of an application scenario of a method for recommending spatiotemporal information according to an exemplary embodiment of the present invention;

[0042] Figure 2 A schematic diagram of a flow chart of a selection prediction method according to an exemplary embodiment of the present invention;

[0043] Figure 3 A flowchart of a method for determining spatiotemporal information when computing power is relatively small according to an exemplary embodiment of the present invention;

[0044] Figure 4 A schematic diagram of a clustering result according to an exemplary embodiment of the present invention;

[0045] Figure 5 A flow chart of a method for determining spatiotemporal information when computing power is large according to an exemplary embodiment of the present invention;

[0046] Figure 6 A schematic diagram of a driving route according to an exemplary embodiment of the present invention;

[0047] Figure 7 Another driving route schematic diagram according to an exemplary embodiment of the present invention;

[0048] Figure 8 A schematic diagram of multiple driving routes according to an exemplary embodiment of the present invention;

[0049] Fig. 9 A schematic diagram of historical travel information according to an exemplary embodiment of the present invention;

[0050] Fig.10 Another schematic diagram of historical travel information according to an exemplary embodiment of the present invention;

[0051] Fig.11 A schematic diagram of a process of pushing a vehicle identification of a target vehicle according to an exemplary embodiment of the present invention;

[0052] Fig.12 A schematic diagram of a spatiotemporal information recommendation device according to an exemplary embodiment of the present invention;

[0053] Fig.13 A schematic diagram of a spatiotemporal information recommendation device according to an exemplary embodiment of the present invention;

[0054] Fig.14 The figure is a schematic diagram of an electronic device according to an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0055] Obviously, the embodiments described below are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without making creative work are within the scope of protection of the present application. In the absence of conflict, the embodiments in the present application and the features in the embodiments can be combined arbitrarily with each other. And, although the logical order is shown in the flow chart, in some cases, the steps shown or described can be performed in an order different from that herein.

[0056] The acquisition, storage, use, and processing of data in the embodiments of this application comply with the relevant provisions of national laws and regulations.

[0057] A method for recommending spatiotemporal information in an embodiment of the present application is described in detail below with reference to the accompanying drawings.

[0058] See also Figure 1 , which is a schematic diagram of an application scenario of a method for recommending spatiotemporal information provided in an embodiment of the present application. The application scenario includes a network 10, an electronic device 20, a database 40, and at least one video and a camera device (the video and camera device 30_1, the video and camera device 30_2, and the video and camera device 30_N in the example in the figure). Furthermore, the application scenario of a method for recommending spatiotemporal information may also include at least one terminal (the terminal 50_1, the terminal 50_2, and the terminal 50_N in the example in the figure), such as Figure 2 shown.

[0059] The video and checkpoint equipment collects the vehicle passing data and uploads it to the electronic device 20; the electronic device 20 analyzes the vehicle passing data to obtain the time and space information to be recommended, and sends it to the target terminal, so that the relevant personnel equipped with the target terminal can arrive at the recommended point before the recommended time; the database 40 is used to store data and programs for the method of determining the time and space information; the electronic device 20, the terminal and the video and checkpoint equipment communicate through the network 10.

[0060] Currently, vehicle passing data is only used for vehicle passing query, vehicle passing statistics and congestion analysis, and there is still more data value to be mined. The embodiment of the present application provides a solution for determining future spatiotemporal information based on the analysis of vehicle passing data.

[0061] The present application embodiment provides a method for recommending spatiotemporal information, which can be divided into the following two methods according to the computing power of the current electronic device:

[0062] If the computing power of the current electronic device is less than the preset computing power, the time-space information to be recommended is determined based on the historical vehicle passing data;

[0063] If the computing power of the current electronic device is not less than the preset computing power, future spatiotemporal information is predicted based on historical vehicle passing data and current vehicle passing data.

[0064] Computing power is the ability of computer equipment to process information, and is the ability of computer hardware and software to work together to perform certain computing requirements. Computing power can specifically include: the computing power of the central processing unit, the computing power of the graphics processing unit, and can also include other things, which are not specifically limited here. If the computing power of current electronic devices can support the prediction of real-time future spatiotemporal information based on real-time acquired vehicle passing data, then only historical vehicle passing data can be used to predict future spatiotemporal information.

[0065] The electronic device can be a device that specifically executes the method for determining spatiotemporal information provided in the embodiment of the present application, or it can be a device that supports the simultaneous execution of multiple methods. The electronic device can be a server, and the embodiment of the present application does not specifically limit this. For devices that execute multiple methods simultaneously. For devices that specifically execute the method for determining spatiotemporal information provided in the embodiment of the present application, as time goes by, more and more vehicle passing data will be acquired, and the computing power resources occupied (for example, computing power resources occupied by data management) will also increase. Therefore, before each execution of the method for determining spatiotemporal information, it is necessary to determine whether the computing power is less than the preset computing power; for devices that support the simultaneous execution of multiple methods, multiple methods may be executed at the same time. Therefore, before each execution of the method for determining spatiotemporal information, it is also necessary to determine whether the computing power is less than the preset computing power to ensure that the appropriate method for determining spatiotemporal information can be selected, thereby allowing the subsequent process to proceed smoothly.

[0066] The implementation method of determining the time-space information to be recommended based on the historical vehicle passing data provided in the embodiment of the present application is as follows: Figure 3 shown.

[0067] S301: Analyze the video in the preset area and the historical vehicle passing data acquired by the checkpoint device to obtain a plurality of spatiotemporal data, and determine the spatiotemporal information corresponding to each spatiotemporal data.

[0068] The video and checkpoint device can be a camera, which is used to collect the vehicle passing data corresponding to the vehicle when the vehicle enters the camera range of the video and checkpoint. The vehicle passing data is the video data of the vehicle collected by the video and checkpoint device. A piece of vehicle passing data can contain data of multiple vehicles. The historical vehicle passing data can be the historical vehicle passing data of the previous month. If the computing power of the current electronic equipment is small, the historical vehicle passing data of the previous week can also be selected for analysis. The embodiments of this application do not make specific limitations.

[0069] Taking the historical vehicle passing data of the previous month as an example, multiple spatiotemporal data including target vehicle images are parsed from the historical vehicle passing data of the previous month, where the target vehicles are vehicles that do not comply with the preset driving rules and vehicles with target identification. For any spatiotemporal data, it may only include images of vehicles that do not comply with the preset driving rules; or only include images of vehicles with target identification; or include images of both vehicles that do not comply with the preset driving rules and vehicles with target identification; or include non-target vehicles and target vehicles.

[0070] Each spatiotemporal data corresponds to a spatiotemporal information, which includes the time and location of the corresponding spatiotemporal data. The time of the spatiotemporal data is the time when the video or camera device collects the spatiotemporal data, and the location of the spatiotemporal data is the location of the video or camera device that collects the spatiotemporal data. For example, the spatiotemporal information corresponding to a spatiotemporal data is (time: 9:00, location A: 20 degrees east longitude, 20 degrees north latitude).

[0071] The spatiotemporal information obtained in the embodiment of the present application is shown in Table 1.

[0072] Table 1

[0073]

[0074] In Table 1, L represents the point, T represents the time, and C represents the spatiotemporal information.

[0075] S302: Clustering each spatiotemporal information according to the similarity between the spatiotemporal information, and determining the risk weight of each category based on the amount of spatiotemporal information in each category.

[0076] In a possible implementation, clustering each spatiotemporal information according to the similarity between each spatiotemporal information includes:

[0077] Convert each space-time information into a corresponding space-time vector.

[0078] Each space-time information contains a time and a point, for example, (9:00, 20 degrees east longitude, 20 degrees north latitude), the corresponding space-time vector is l represents the point, t represents the time, the first row represents the vector corresponding to the time, and the second row represents the vector corresponding to the point.

[0079] The space-time information corresponding to each space-time vector is clustered according to the distance between each space-time vector.

[0080] After converting all the space-time information into space-time vectors, cluster the space-time vectors. Specifically, the following steps can be used:

[0081] (1) Initialization.

[0082] Randomly select k space-time vectors from each space-time vector as the initial clustering center: {u1,u2,u3,...,u k}, where u1 is the first cluster center, u2 is the second cluster center, u3 is the third cluster center, and u k is the first cluster center.

[0083] (2) Cluster each space-time vector.

[0084] Clustering is performed according to the cluster centers in (1) above. Determine each space-time vector x i =(l i ,t i ),i=1,2,...,n are the distances to k cluster centers respectively, d ij =||x i -u j || 2 , if the distance between the i-th space-time vector and the j-th cluster center is the shortest, then the i-th space-time vector is classified into the j-th category.

[0085] Among them, the above x i is the i-th space-time vector, l i is the time vector in the i-th space-time vector, t i represents the point vector in the i-th space-time vector; d ij is the i-th space-time vector x i To the jth cluster center u j distance.

[0086] For example, the i-th space-time vector The jth cluster center is ||x i -u j || 2 =(-3) 2 =9, i.e. the i-th space-time vector x i To the jth cluster center u j The distance is 9. If the i-th space-time vector x i If the distance from other cluster centers is greater than 9, then x i Belongs to the jth category.

[0087] (3) Update cluster centers.

[0088] When the number of iterations does not reach the preset number, the cluster center is updated and a new round of clustering is performed. Update the cluster centers, and then perform a new round of clustering based on the updated cluster centers.

[0089] Among them, u j is the j-th cluster center before the update, m is the number of space-time vectors belonging to the j-th class before the update, and x represents the space-time vector belonging to the j-th class before the update.

[0090] For example, the jth class contains the space-time vector and Initial cluster centers The updated cluster center is

[0091] When the updated cluster center is different from each space-time vector (such as x1, x2 and x3), then x3 needs to be used as the space-time vector to be clustered for a new round of clustering process.

[0092] (4) Output clustering results.

[0093] When the stop updating condition is met, the update is stopped and the clustering result is output. The stop condition update condition can be that the update number is reached, or the current k cluster centers are the same as the k cluster centers before the update.

[0094] After obtaining k categories and the space-time information corresponding to the space-time vectors contained in each category, the risk weight corresponding to each category is determined according to the amount of space-time information in each category. The risk weight is a preset value, and the number of risk weights is the same as the number of categories. The more space-time information there is, the more frequently the target vehicle appears at the point and time of this category, and the greater the corresponding risk weight. Figure 4 The result after clustering is shown, which divides the spatiotemporal information into k categories. Each category contains the amount of spatiotemporal information and the corresponding weight. For example, the amount of spatiotemporal information in the first category is C1 and the weight is w1.

[0095] For example, after clustering, three categories are obtained, and the preset three risk weights are 1.3, 1.1 and 0.9 respectively. The amount of spatiotemporal information in the first category is 1000, and the corresponding risk weight is 1.3; the amount of spatiotemporal information in the second category is 900, and the corresponding risk weight is 1.1; the amount of spatiotemporal information in the third category is 700, and the corresponding risk weight is 0.9.

[0096] S303: Based on the risk weight of each category, determine the spatiotemporal information to be recommended from the spatiotemporal information.

[0097] According to the embodiment in S302 above, the risk weight of the first category is 1.3, the risk weight of the second category is 1.1, and the risk weight of the third category is 0.9. The spatiotemporal information in the first category with a risk weight of 1.3 is determined as the spatiotemporal information to be recommended. Before sending it to the terminal, the spatiotemporal information in the first category can be further sorted.

[0098] If the 100 spatiotemporal information included in the first category have the same point and time, or the same point and similar time, the 100 spatiotemporal information can be further integrated into one spatiotemporal information. For example, the point corresponding to the 100 spatiotemporal information is point A (20 degrees east longitude, 20 degrees north latitude), and the corresponding time is a time range of 9:00 to 10:00, and the spatiotemporal information (time: 9:00 to 10:00, point A) can be used as a piece of spatiotemporal information to be recommended.

[0099] The implementation method provided in the embodiment of the present application for predicting future spatiotemporal information based on historical vehicle passing data and current vehicle passing data is as follows: Figure 5 shown.

[0100] S501: Analyze the video in the preset area and the historical vehicle passing data acquired by the checkpoint device to obtain the historical travel information of the target vehicle.

[0101] The target vehicles include vehicles that do not comply with preset driving rules and vehicles with target identification; each target vehicle corresponds to a historical travel information. Determining the historical travel information of any target vehicle includes:

[0102] Filtering out the historical vehicle passing data of any one of the target vehicles from the historical vehicle passing data;

[0103] The historical vehicle passing data of any one of the target vehicles is analyzed to obtain the historical travel information of any one of the target vehicles.

[0104] The historical vehicle passing data includes the target vehicle passing data and the non-target vehicle passing data. In order to predict the spatiotemporal information, the embodiment of the present application selects the target vehicle passing data. Since each target vehicle corresponds to a historical travel information, the embodiment of the present application further distinguishes the historical vehicle passing data of each target vehicle.

[0105] For any target vehicle, historical travel information is determined using historical vehicle data, including:

[0106] Determine at least one driving route of any target vehicle according to the historical vehicle passing data of any target vehicle;

[0107] For any driving route, according to the driving direction of any driving route, the number of times each adjacent point on any driving route corresponds is counted, wherein the number of times the adjacent point corresponds is the number of times any target vehicle reaches another point from one of the adjacent points according to the driving direction;

[0108] For any adjacent point, based on the number of times corresponding to any adjacent point, determine the transfer probability of any target vehicle from one of the adjacent points to another point according to the driving direction;

[0109] The historical travel information of any one of the target vehicles is determined based on the transition probability between adjacent points on the at least one route.

[0110] like Figure 6 As shown, a driving route of any target vehicle provided in an embodiment of the present application is provided, and the driving route includes the driving direction of any target vehicle, the points passed by any target vehicle, and the time of arrival at each point.

[0111] The driving direction corresponding to this driving route is A→B→C→D. The time when point A is passed is 9:00, the time when point B is passed is 9:05, the time when point C is passed is 9:15, and the time when point D is passed is 9:30.

[0112] The time to reach each point can be a specific time point or a time range representing the time required to reach from one point to another, such as Figure 7 The time from point A to point B is 5 to 6 minutes, the time from point B to point C is 9 to 11 minutes, and the time from point C to point D is 13 to 15 minutes.

[0113] For any target vehicle corresponding to multiple driving routes, perform the following operations.

[0114] (1) Delete the driving routes whose length is not greater than the preset length.

[0115] The length of the driving route can be measured by the number of points included, wherein the points included in the driving route are the points passed by any target vehicle. The fewer the number of points included in the driving route, the shorter the length of the driving route.

[0116] For example, the preset length is 2, that is, the number of points is 2. According to the historical passing data of any target vehicle, a driving route A→B with a length of 2 is obtained, indicating that the video or camera equipment that collected this historical passing data may have wrong shots or missed shots, and the data of this driving route is deleted.

[0117] (2) Merge two driving routes whose similarity is less than a preset value.

[0118] The method for determining the similarity between two driving routes provided in the embodiment of the present application is to determine the edit distance between one driving route and another driving route. Other determination methods may also be used, and the present application does not specifically limit them. Among them, the edit distance includes three operations: insertion, deletion, and replacement.

[0119] For example, if the first driving route is A→B→C→D and the second driving route is A→B→X→Y→D, the editing process is as follows:

[0120] Replace point C in the first driving route with point X, and obtain the driving route after the first edit as A→B→X→D; then insert point Y into the driving route after the first edit, and obtain the second driving route A→B→X→Y→D. The editing distance from the first driving route to the second driving route is 2. When it is determined that point X can be reached directly from point C, the two driving routes can be further merged to obtain a merged driving route of A→B→C→X→Y→D; if it is found that the number of times any target vehicle passes through point C is small based on the historical vehicle passing data of any target vehicle, the merged driving route can also be determined to be A→B→X→Y→D. The embodiment of the present application does not specifically limit the merging method of driving routes. According to the analysis results of the historical vehicle passing data, some driving routes that need to be merged can be selected from multiple driving routes to be merged whose similarity is less than a preset value; after obtaining the merged driving route, it can also be further determined whether the merged driving route needs to continue to be merged with other driving routes.

[0121] After deleting and merging the obtained driving routes, we get Figure 8 At least one driving route corresponding to any target vehicle shown includes: driving route A→B→C→E, driving route A→B→E, driving route A→B→D→F→E, driving route A→G→B→C→E, driving route A→G→B→D→F→E, driving route A→G→B→E and driving route A→D→F→E.

[0122] When counting the number of times one adjacent point reaches another, it is necessary to add up the number of times on each driving route. For example, for the total number of times from point B to point C, add the number of times from point B to point C in driving route A→B→C→E (100) and the number of times from point B to point C in driving route A→G→B→C→E (50), and the total number of times from point B to point C is 150.

[0123] For a point, its corresponding adjacent points may include multiple points. For example, according to the driving direction, the points adjacent to point B include point C, point D and point E. The transfer probability of point B reaching point C, point D and point E respectively can be determined according to the number of times point B reaches point C, point D and point E respectively.

[0124] For example, the total number of times point B reaches point C, point D and point E is 1000, the number of times point B reaches point C is 300, and the transfer probability is 300 / 1000=0.3; the number of times point B reaches point D is 200, and the transfer probability is 200 / 1000=0.2; the number of times point B reaches point E is 500, and the transfer probability is 500 / 1000=0.5.

[0125] The obtained transfer probability and driving route are used as the historical travel information of any target vehicle. Fig. 9 shown. Fig. 9 The historical travel information of any target vehicle shown in includes: the travel direction on each travel route, such as A→B, the points passed on each travel route, such as A, B, C, etc., the time to reach each point, such as t1, t2, etc., and the transition probability of one point to another in each adjacent point, such as 0.5, 0.3, etc. marked on the arrow.

[0126] S502: If the spatiotemporal data of any target vehicle is obtained through the video and vehicle passing data currently acquired by the toll station device in the preset area, the current spatiotemporal information of the any target vehicle is determined based on the spatiotemporal data of the any target vehicle.

[0127] If the computing power of the electronic device is not less than the preset computing power value, the future spatiotemporal information can be predicted based on the historical vehicle passing data and the current vehicle passing data. The analysis of the historical vehicle passing data is as in the implementation method of S501; the current vehicle passing data is the image data of some or all of the target vehicles, and the spatiotemporal information corresponding to each target vehicle can be further obtained based on the image data.

[0128] When the current passing data of the target vehicle is obtained through the video and the camera device, the subsequent process of predicting future spatiotemporal information continues, otherwise the prediction process ends. In order to avoid unnecessary calculation processes, the judgment of whether the current passing data of the target vehicle is obtained can be placed before parsing the historical passing data.

[0129] S503: Predicting the future space-time information of any one of the target vehicles based on the historical travel information of the target vehicle and the current space-time information.

[0130] When predicting the future spatiotemporal information of any target vehicle, it is first necessary to search for target spatiotemporal information that matches the current spatiotemporal information in the historical travel information of any target vehicle. If not found, search for target spatiotemporal information that matches the current spatiotemporal information in the historical travel information of other target vehicles except the any target vehicle. Then, based on the transfer probability of the point corresponding to the queried target spatiotemporal information to reach the next point, and the time contained in the target spatiotemporal information, predict the future spatiotemporal information of any target vehicle.

[0131] For different target vehicles, in order to avoid on-site inspection, points or driving routes that are not subject to on-site inspection may be selected; in addition, if the target vehicle wants to reach the target point, there may be points that must be passed, for example, Fig. 9 In the example, if you want to reach point F, you must pass through point D. Based on this, there may be similarities between the historical travel information of different target vehicles. Therefore, when there is no target spatiotemporal information matching the current spatiotemporal information in the historical travel information of any target vehicle, you can further randomly select a historical travel information from the historical travel information of other target vehicles except any target vehicle, and query the target spatiotemporal information matching the current spatiotemporal information until the target spatiotemporal information matching the current spatiotemporal information is found.

[0132] If target spatiotemporal information matching the current spatiotemporal information is not found in the historical travel information of other target vehicles except any one target vehicle, the process of predicting future spatiotemporal information is terminated.

[0133] For example, the current spatiotemporal information of any target vehicle is (time: 9:00, location: B). Fig. 9 The matching target spatiotemporal information is found in the historical travel information. Fig. 9 Based on the historical travel information in the data, it is determined that the probability of going from point B to point E is the largest (0.5), and any target vehicle will arrive at point E at 9:00+t5 in the future. Due to road conditions, human factors and other factors, the current time of any target vehicle may not completely match the time in the historical travel information. For example, the current arrival time at point B is 9:00, and the arrival time at point B in the historical travel information is 8:00, 8:10, etc. At this time, the time range required to reach point C from point B can be calculated based on the historical travel information, and the future time can be predicted using this time range.

[0134] For example, Fig. 9 Shown is the historical travel information of any target vehicle. Fig.10The historical travel information of other target vehicles is shown. If the current spatiotemporal information of any target vehicle is determined to be (time: 9:00, location: E), then Fig. 9 The historical travel information shown cannot obtain the spatiotemporal information from point E to other points. Fig.10 By querying the historical travel information shown, it is found that the maximum probability of going from point E to point K is 0.6, and the predicted future spatiotemporal information is determined to be (time: 9:00+t3, point: K).

[0135] In addition to recommending future spatiotemporal information, the embodiment of the present application also provides a solution for recommending vehicle identification. For any target vehicle, the specific steps of recommending vehicle identification are as follows: Fig.11 shown.

[0136] S1101: Determine whether the current passing data of any target vehicle is obtained. If so, execute S1102, otherwise end the process. Any target vehicle may be a vehicle that does not comply with the preset driving rules (such as an overloaded vehicle, a red light runner), or a vehicle with a target identification. The vehicle with the target identification is a vehicle in a preset list (for example, a vehicle that has not been inspected annually or has not been scrapped overdue).

[0137] S1102: Determine whether the target vehicle has a target identifier, that is, determine whether the identifier of any target vehicle is in the preset list. If so, execute S1104, otherwise execute S1103;

[0138] S1103: Determine whether any target vehicle meets the preset driving rules. If yes, end the process, otherwise execute S1104;

[0139] S1104: taking the identification (license plate number) of any target vehicle as the identification to be recommended, and then terminating the process.

[0140] By analyzing the historical vehicle passing data, the target vehicle identification set can be obtained. According to the video and the camera device, if an identification in the identification set is currently obtained, it is necessary to further determine whether the vehicle corresponding to the identification is a vehicle in the preset list, or whether it violates the preset driving rules at present. If it is a vehicle in the preset list, or whether it violates the preset driving rules at present, the identification (license plate number) of the vehicle is pushed to the target terminal for relevant personnel to check.

[0141] The method for predicting future space-time information provided by the embodiment of the present application can accurately and timely push the future space-time information to the target terminal, so that the relevant personnel equipped with the target terminal can arrive at the recommended point before the recommended time, stop the target vehicle from continuing to drive, and effectively prevent the occurrence of traffic accidents; moreover, the historical vehicle passing data of vehicles that do not meet the preset driving rules and the historical vehicle passing data of vehicles with target identification are combined to predict future space-time information, so that the prediction result is more accurate.

[0142] Based on the same inventive concept, the present application embodiment also provides a spatiotemporal information recommendation device, such as Fig.12 As shown, the device comprises:

[0143] The parsing module 1201 is used to parse the video and historical vehicle passing data acquired by the checkpoint device in the preset area to obtain multiple spatiotemporal data, and determine the spatiotemporal information corresponding to each spatiotemporal data, wherein each spatiotemporal data includes an image of a vehicle that does not comply with the preset driving rules and / or an image of a vehicle with a target identification, and the spatiotemporal information includes the time and location of the corresponding spatiotemporal data, the time of the spatiotemporal data is the time when the video and checkpoint device collect the spatiotemporal data, and the location of the spatiotemporal data is the location of the video and checkpoint device that collect the spatiotemporal data;

[0144] A clustering module 1202 is used to cluster the spatiotemporal information according to the similarity between the spatiotemporal information, and determine the risk weight of each category based on the amount of spatiotemporal information in each category;

[0145] The recommendation module 1203 is used to determine the spatiotemporal information to be recommended from the spatiotemporal information based on the risk weight of each category.

[0146] In a possible implementation, Fig.13 As shown, the device also includes a computing power determination module 1204, which is used to parse the video located in the preset area and the historical vehicle passing data acquired by the checkpoint device, and determine that the computing power of the current electronic device is less than the preset computing power value before obtaining multiple spatiotemporal data.

[0147] In a possible implementation, the parsing module 1201 is used to parse the video in the preset area and the historical vehicle passing data acquired by the checkpoint device when the computing power of the current electronic device is not less than the preset computing power value, and obtain the historical travel information of the target vehicle, wherein the target vehicle includes a vehicle that does not comply with the preset driving rules and a vehicle with a target identification;

[0148] If the spatiotemporal data of any target vehicle is obtained through the video and the vehicle passing data currently acquired by the checkpoint device in the preset area, the current spatiotemporal information of the any target vehicle is determined based on the spatiotemporal data of the any target vehicle;

[0149] Based on the historical travel information of the target vehicle and the current spatiotemporal information, the future spatiotemporal information of any one of the target vehicles is predicted.

[0150] In a possible implementation, the parsing module 1201 is used to filter out the historical vehicle passing data of any target vehicle from the historical vehicle passing data;

[0151] The historical vehicle passing data of any one of the target vehicles is analyzed to obtain the historical travel information of any one of the target vehicles.

[0152] In a possible implementation, the parsing module 1201 is used to determine at least one driving route of any one of the target vehicles based on the historical vehicle passing data of any one of the target vehicles, wherein each first driving route includes the driving direction of any one of the target vehicles, the points passed by any one of the target vehicles, and the time of arrival at each point;

[0153] For any driving route, according to the driving direction of the any driving route, the number of times corresponding to each adjacent point on the any driving route is counted, wherein the number of times corresponding to the adjacent point is the number of times the any target vehicle reaches another point from one of the adjacent points according to the driving direction;

[0154] For any adjacent point, based on the number of times corresponding to the any adjacent point, determine the transfer probability of the any target vehicle from one of the adjacent points to another point according to the driving direction;

[0155] The historical travel information of any one of the target vehicles is determined based on the transition probability between adjacent points on the at least one route.

[0156] In a possible implementation, the parsing module 1201 is used to query the target spatiotemporal information matching the current spatiotemporal information in the historical travel information of any target vehicle for any target vehicle, and if no target information is found, query the target spatiotemporal information matching the current spatiotemporal information in the historical travel information of other target vehicles other than the any target vehicle;

[0157] According to the transfer probability of the point corresponding to the queried target spatiotemporal information to the next point, and the time contained in the target spatiotemporal information, the future spatiotemporal information of any target vehicle is predicted. In a possible implementation, the clustering module 1202 is used to convert each spatiotemporal information into a corresponding spatiotemporal vector, and cluster the spatiotemporal information corresponding to each spatiotemporal vector according to the distance between the spatiotemporal vectors.

[0158] Based on the same inventive concept, an embodiment of the present application further provides an electronic device, the device comprising:

[0159] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute any of the spatiotemporal information recommendation methods in the above-mentioned embodiments.

[0160] Refer to the following Fig.14 The electronic device 130 according to this embodiment of the present application is described. Fig.14 The electronic device 130 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0161] like Fig.14 As shown, the electronic device 130 is in the form of a general electronic device. The components of the electronic device 130 may include but are not limited to: the at least one processor 131, the at least one memory 132, and a bus 133 connecting different system components (including the memory 132 and the processor 131).

[0162] The processor 131 is used to read and execute instructions in the memory 132, so that the at least one processor can execute a spatiotemporal information recommendation method provided by the above embodiment.

[0163] Bus 133 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a processor, or a local bus using any of a variety of bus architectures.

[0164] The memory 132 may include a readable medium in the form of a volatile memory, such as a random access memory (RAM) 1321 and / or a cache memory 1322 , and may further include a read-only memory (ROM) 1323 .

[0165] The memory 132 may also include a program / utility 1325 having a set (at least one) of program modules 1324, such program modules 1324 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0166] The electronic device 130 may also communicate with one or more external devices 134 (e.g., keyboards, pointing devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 130, and / or communicate with any device that enables the electronic device 130 to communicate with one or more other electronic devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 135. Furthermore, the electronic device 130 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 136. As shown, the network adapter 136 communicates with other modules for the electronic device 130 via a bus 133. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 130, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0167] In some possible implementations, various aspects of a spatiotemporal information recommendation method provided by the present application may also be implemented in the form of a program product, which includes program code. When the program product is run on a computer device, the program code is used to enable the computer device to execute the steps of a spatiotemporal information recommendation method according to various exemplary embodiments of the present application described above in this specification.

[0168] In addition, the present application also provides a computer-readable storage medium, wherein the computer storage medium stores a computer program, wherein the computer program is used to enable a computer to execute any of the methods described in the above embodiments.

[0169] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0170] 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.

[0171] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0172] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and 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 method for recommending spatiotemporal information, characterized in that: Applied to electronic equipment, the method comprises: Parse the historical vehicle passing data acquired by the video and the checkpoint device in the preset area to obtain a plurality of spatiotemporal data, and determine the spatiotemporal information corresponding to each spatiotemporal data, wherein each spatiotemporal data includes an image of a vehicle that does not comply with the preset driving rules and / or an image of a vehicle with a target identification, and the spatiotemporal information includes the time and point of the corresponding spatiotemporal data, the time of the spatiotemporal data is the time when the video and the checkpoint device collect the spatiotemporal data, and the point of the spatiotemporal data is the position of the video and the checkpoint device that collect the spatiotemporal data; Clustering the spatiotemporal information according to the similarity between the spatiotemporal information, and determining the risk weight of each category based on the amount of spatiotemporal information in each category; Based on the risk weight of each category, the spatiotemporal information to be recommended is determined from the spatiotemporal information.

2. The method according to claim 1, characterized in that Before analyzing the video in the preset area and the historical vehicle passing data acquired by the checkpoint device to obtain the plurality of spatiotemporal data, the method further includes: Determine that the current computing power of the electronic device is less than a preset computing power value.

3. The method according to claim 2, characterized in that The method further comprises: If the computing power of the current electronic device is not less than the preset computing power value, the historical vehicle passing data acquired by the video and the checkpoint device in the preset area is analyzed to obtain the historical travel information of the target vehicle, wherein the target vehicle includes a vehicle that does not comply with the preset driving rules and a vehicle with a target identification; If the spatiotemporal data of any target vehicle is obtained through the video and the vehicle passing data currently acquired by the checkpoint device in the preset area, the current spatiotemporal information of the any target vehicle is determined based on the spatiotemporal data of the any target vehicle; Based on the historical travel information of the target vehicle and the current spatiotemporal information, the future spatiotemporal information of any one of the target vehicles is predicted.

4. The method according to claim 3, characterized in that The analysis of the video in the preset area and the historical vehicle passing data acquired by the checkpoint device to obtain the historical travel information of the target vehicle includes: For any target vehicle, filter out the historical vehicle passing data of the target vehicle from the historical vehicle passing data; The historical vehicle passing data of any one of the target vehicles is analyzed to obtain the historical travel information of any one of the target vehicles.

5. The method according to claim 4, characterized in that The parsing of the historical vehicle passing data of any one of the target vehicles to obtain the historical travel information of any one of the target vehicles includes: Determine at least one driving route of any one of the target vehicles according to the historical vehicle passing data of any one of the target vehicles, wherein each first driving route includes the driving direction of any one of the target vehicles, the points passed by any one of the target vehicles, and the time of arrival at each point; For any driving route, according to the driving direction of the any driving route, the number of times corresponding to each adjacent point on the any driving route is counted, wherein the number of times corresponding to the adjacent point is the number of times the any target vehicle reaches another point from one of the adjacent points according to the driving direction; For any adjacent point, based on the number of times corresponding to the any adjacent point, determine the transfer probability of the any target vehicle from one of the adjacent points to another point according to the driving direction; The historical travel information of any one of the target vehicles is determined based on the transition probability between adjacent points on the at least one route.

6. The method according to claim 5, characterized in that The predicting of the future spatiotemporal information of any one of the target vehicles based on the historical travel information of the target vehicle and the current spatiotemporal information includes: For any target vehicle, query the historical travel information of any target vehicle for target spatiotemporal information that matches the current spatiotemporal information; if no target information is found, query the historical travel information of other target vehicles other than the any target vehicle for target spatiotemporal information that matches the current spatiotemporal information; The future spatiotemporal information of any target vehicle is predicted based on the transfer probability of the point corresponding to the queried target spatiotemporal information to the next point and the time included in the target spatiotemporal information.

7. The method according to any one of claims 1 to 6, characterized in that: Clustering the various spatiotemporal information according to the similarity between the various spatiotemporal information includes: Each piece of space-time information is converted into a corresponding space-time vector, and the space-time information corresponding to each space-time vector is clustered according to the distance between the space-time vectors.

8. A spatiotemporal information recommendation device, characterized in that: The device comprises: A parsing module is used to parse the video in a preset area and the historical vehicle passing data acquired by the checkpoint device to obtain a plurality of spatiotemporal data, and determine the spatiotemporal information corresponding to each spatiotemporal data, wherein each spatiotemporal data includes an image of a vehicle that does not comply with the preset driving rules and / or an image of a vehicle with a target identification, and the spatiotemporal information includes the time and location of the corresponding spatiotemporal data, the time of the spatiotemporal data is the time of the historical vehicle passing data acquired by the video and the checkpoint device, and the location of the spatiotemporal data is the location of the video and the checkpoint device that collected the spatiotemporal data; A clustering module, for clustering the spatiotemporal information according to the similarity between the spatiotemporal information, and determining the risk weight of each category based on the amount of spatiotemporal information in each category; A recommendation module is used to determine the spatiotemporal information to be recommended from the spatiotemporal information based on the risk weight of each category.

9. An electronic device, characterized in that: The device comprises: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor so that the at least one processor can execute the method as described in any one of claims 1-7.

10. A computer storage medium, characterized in that: The computer storage medium stores a computer program, and the computer program is used to enable a computer to execute the method according to any one of claims 1 to 7.

Images