Vehicle passing behavior analysis method, device and computer storage medium

Abstract

The present application provides a method, device, and computer storage medium for analyzing vehicle passing behaviors, mainly including: responding to an analysis instruction of a target vehicle, continuously acquiring driving data of the target vehicle within a preset analysis time limit; analyzing the vehicle passing behaviors of the target vehicle according to the driving data to obtain an analysis result of normal or abnormal vehicle passing behaviors; and if the analysis result is abnormal vehicle passing behaviors and new driving data is acquired within the preset analysis time limit, re-analyzing the vehicle passing behaviors of the target vehicle according to the new driving data to update the analysis result. Thereby, the present application accurately analyzes whether the vehicle passing behaviors are normal, and can effectively reduce the occurrence of phenomena such as highway toll evasion.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of intelligent transportation, and particularly to a method, device, and computer storage medium for analyzing vehicle passing behaviors. Background Art

[0002] With the continuous expansion of the highway road network, some illegal vehicles use various forms to evade tolls. The forms of toll evasion are diverse and highly concealed, disrupting the normal order of highway toll collection work. Most current inspection methods conduct abnormal inspections through offline data and generally can only analyze data on a daily basis. It is impossible to immediately determine whether a vehicle is abnormal after it exits the highway, resulting in difficulties in post-event inspection, making it difficult to trace, and the low illegal cost of toll-evading vehicles causing huge economic losses.

[0003] In view of this, how to provide a vehicle passing behavior analysis technology to improve the serious problem of highway toll evasion is the technical problem to be solved by the present application. Summary of the Invention

[0004] In view of the above problems, the present application provides a method, device, and computer storage medium for analyzing vehicle passing behaviors, which can accurately analyze whether the vehicle passing behavior is abnormal to reduce the occurrence of abnormal highway toll payment.

[0005] The first aspect of the present application provides a method for analyzing vehicle passing behaviors, which includes: responding to an analysis instruction of a target vehicle, continuously obtaining driving data of the target vehicle within a preset analysis time limit; analyzing the vehicle passing behavior of the target vehicle according to the driving data to obtain an analysis result of normal or abnormal vehicle passing behavior; and if the analysis result is abnormal vehicle passing behavior and new driving data is obtained within the preset analysis time limit, re-analyzing the vehicle passing behavior of the target vehicle according to the new driving data to update the analysis result.

[0006] The second aspect of the present application provides a computer storage medium, in which instructions for executing the steps in the method described in the first aspect above are stored.

[0007] A third aspect of the present application provides a vehicle passing behavior analysis device, which includes: a data acquisition module, configured to continuously acquire driving data of the target vehicle within a preset analysis time limit in response to an analysis instruction of the target vehicle; a behavior analysis module, configured to analyze the vehicle passing behavior of the target vehicle according to the driving data, and obtain an analysis result indicating that the vehicle passing behavior is normal or abnormal; wherein, if the analysis result of the behavior analysis module indicates that the vehicle passing behavior is abnormal, and the data acquisition module acquires new driving data within the preset analysis time limit, the behavior analysis module is provided to re-analyze the vehicle passing behavior of the target vehicle according to the new driving data, so as to update the analysis result.

[0008] In summary, the vehicle passing behavior analysis method, device and computer storage medium according to the embodiments of the present application can continuously acquire the driving data of the target vehicle within a preset analysis time limit, and analyze whether the passing behavior of the target vehicle is normal based on this, thereby reducing the occurrence of abnormal highway tolls. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments recorded in the embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0010] Figure 1 It is a schematic flowchart of the vehicle passing behavior analysis method according to the first embodiment of the present application.

[0011] Figure 2 It is a schematic flowchart of the vehicle passing behavior analysis method according to the second embodiment of the present application.

[0012] Figure 3 It is a schematic flowchart of the vehicle passing behavior analysis method according to the third embodiment of the present application.

[0013] Figure 4 It is a schematic flowchart of the vehicle passing behavior analysis method according to the fourth embodiment of the present application.

[0014] Figure 5 It is a schematic flowchart of the vehicle passing behavior analysis method according to the fifth embodiment of the present application.

[0015] Figure 6 It is a schematic flowchart of the vehicle passing behavior analysis method according to the sixth embodiment of the present application.

[0016] Figure 7 It is a schematic flowchart of the vehicle passing behavior analysis method according to the seventh embodiment of the present application.

[0017] Figure 8 It is a schematic flow chart of the vehicle passing behavior analysis device according to the ninth embodiment of the present application.

[0018] Element label

[0019] 800: Vehicle passing behavior analysis device; 802: Data acquisition module; 804: Behavior analysis module. Detailed implementation manners

[0020] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art shall fall within the protection scope of the embodiments of the present application.

[0021] The following will describe each embodiment of the present application in detail with reference to the accompanying drawings.

[0022] The first embodiment

[0023] Figure 1 It shows a schematic flow chart of the vehicle passing behavior analysis method according to the first embodiment of the present application. As shown in the figure, this embodiment mainly includes the following steps:

[0024] Step S102, in response to the analysis instruction of the target vehicle, continuously acquire the driving data of the target vehicle within a preset analysis time limit.

[0025] Optionally, the driving data of the target vehicle may include at least one of the driving-in data of the target vehicle entering the target road network, the driving-out data of the target vehicle leaving the target road network, and the passing data of the target vehicle driving on the target road network.

[0026] Optionally, the target road network may include a highway network.

[0027] Optionally, the driving-in data and driving-out data of the target vehicle can be obtained through the entrance and exit toll gates set in the target road network, and the passing data of the target vehicle can be obtained through each image capture device set on each section of the target road network.

[0028] For example, when the target vehicle enters the highway network, data such as the identification information of the target vehicle (such as license plate number), the driving-in node (such as highway entrance), the driving-in time, the vehicle type specification (such as car, bus, truck, etc.), and the vehicle load can be obtained to form the driving-in data of the target vehicle.

[0029] For example, when the target vehicle leaves the highway network, data such as the identification information of the target vehicle (e.g., license plate number), the departure node (e.g., highway exit), the departure time, and the transaction fees generated by passing through the highway network can be obtained to form the departure data of the target vehicle.

[0030] For example, during the process of the target vehicle traveling on the highway network, data such as the captured image of the target vehicle, the capture time, and the capture location can be obtained through the gantry capture devices set in the highway network to form the passing data of the target vehicle.

[0031] In this embodiment, the Hadoop Distributed File System (HDFS) can be used to store the driving data of the target vehicle.

[0032] Optionally, the analysis instruction of the target vehicle can include automatic triggering that meets the triggering conditions and / or manual triggering based on user input.

[0033] In one embodiment, the analysis instruction can be triggered when the departure data of the target vehicle is obtained. For example, when the exit boundary gantry of the highway network captures an image of the target vehicle, the analysis instruction of the target vehicle is automatically triggered.

[0034] In another embodiment, the data reception time limit of the target vehicle can be dynamically updated according to the entry data or passing data of the target vehicle. If no new driving data is received within the data reception time limit, the analysis instruction is triggered. For example, assuming that the set data reception time limit is 24 hours, if 4 hours after receiving the entry data or passing data of the target vehicle, new passing data of the target vehicle is received, then based on the update time of the new passing data, the data reception time limit of the target vehicle is refreshed to 24 hours. If no passing data or departure data of the target vehicle is received 24 hours after receiving the latest passing data, the analysis instruction of the target vehicle is automatically triggered.

[0035] In yet another embodiment, the analysis instruction of the target vehicle can also be triggered by the staff manually inputting an instruction.

[0036] Step S104, analyze the vehicle passing behavior of the target vehicle according to the driving data, and obtain the analysis result that the vehicle passing behavior is normal or the vehicle passing behavior is abnormal.

[0037] Optionally, according to the entry data, departure data, and passing data of the target vehicle, the driving trajectory of the target vehicle in the target road network can be analyzed, and based on this, the analysis result that the vehicle passing behavior is normal or the vehicle passing behavior is abnormal can be obtained.

[0038] Optionally, based on the driving data of the target vehicle, the vehicle model specifications, transaction fees, and driving mileage of the target vehicle can be obtained, and based on the vehicle model specifications, transaction fees, and driving mileage, the analysis result of normal or abnormal vehicle passing behavior can be obtained.

[0039] Specifically, based on the driving-in data of the target vehicle, the vehicle model specifications of the target vehicle can be obtained. For example, based on the vehicle image of the target vehicle captured by the highway entrance gantry, the vehicle model specifications of the target vehicle can be determined. Based on the driving data of the target vehicle, each driving segment of the target vehicle in the target road network can be obtained, and based on the mileage of each driving segment corresponding to each driving segment of the target vehicle, the driving mileage of the target vehicle in the target road network can be determined; in addition, based on the driving-out data of the target vehicle, the transaction fees generated by the target vehicle passing through the target road network can be obtained, and then based on the vehicle model specifications and driving mileage of the target vehicle, the passing fees of the target vehicle can be estimated; and by comparing whether the passing fees match the actually generated transaction fees, it can be analyzed whether there is any abnormality in the vehicle passing behavior of the target vehicle.

[0040] Step S106, if the analysis result is that the vehicle passing behavior is abnormal and new driving data is obtained within the preset analysis time limit, re-analyze the vehicle passing behavior of the target vehicle based on the new driving data to update the analysis result.

[0041] In this embodiment, since each piece of driving data of the target vehicle is not pushed serially, that is, each piece of driving data is not pushed in the order of acquisition time, there may be a situation where the driving-out data of the target vehicle is received but its driving-in data or passing data is not received. In view of this, in this application, by setting a preset analysis time limit, if the analysis result is that the vehicle passing behavior is abnormal, it continues to detect whether there is new driving data input. If so, re-analyze the vehicle passing behavior of the target vehicle based on the new driving data, so as to avoid the situation where the analysis result is incorrect due to incomplete reception of the driving data of the target vehicle.

[0042] In summary, the vehicle passing behavior analysis method of the embodiment of this application analyzes whether there is any abnormality in the passing behavior of the target vehicle in the target road network by receiving the driving data of the target vehicle within the preset analysis time limit. If the analysis result is that the vehicle passing behavior is abnormal and new driving data is obtained within the preset analysis time limit, the vehicle passing behavior of the target vehicle can be re-analyzed based on the new driving data to update the analysis result. Accordingly, this application can avoid the problem that the analysis result is incorrect due to incomplete reception of the driving data of the target vehicle.

[0043] Second Embodiment

[0044] Figure 2The flowchart of the vehicle passing behavior analysis method according to the second embodiment of the present application is shown. This embodiment mainly shows the specific implementation of the above step S104. As shown in the figure, this embodiment mainly includes the following steps:

[0045] Step S202, obtain the driving trajectory of the target vehicle in the target road network according to the driving-in data, driving-out data, and passing data of the target vehicle.

[0046] In this embodiment, the driving-in position of the target vehicle relative to the target road network can be obtained according to the driving-in data, the driving-out position of the target vehicle relative to the target road network can be obtained according to the driving-out data, and the driving section where the target vehicle passes through the target road network can be obtained according to the passing data, so as to determine the driving trajectory of the target vehicle in the target road network according to the driving-in position, driving section, and driving-out position of the target road network.

[0047] Step S204, segment the driving trajectory according to each road network node in the target road network to obtain each driving segment of the target vehicle.

[0048] In this embodiment, each road network node may include an end node corresponding to a single road segment in the target road network and an intersection node corresponding to multiple road segments.

[0049] Specifically, the driving trajectory can be segmented according to each road network node to obtain each driving segment of the target vehicle, and the starting node and ending node corresponding to each driving segment can be determined to form the starting and ending nodes of each driving segment.

[0050] Optionally, the segmentation can be performed on the driving trajectory based on two adjacent road network nodes in the target network. In this state, each obtained driving segment generally only includes two road network nodes, that is, the starting node and the ending node located at both ends of the driving segment (which can also be called the starting and ending nodes of the driving segment).

[0051] Optionally, the segmentation can be performed on the driving trajectory based on two non-adjacent road network nodes in the target network. In this state, each obtained driving segment may include at least three road network nodes, that is, in addition to the starting node and the ending node located at both ends of the driving segment, at least one intermediate node located in the middle position of the driving segment may also be included.

[0052] Step S206, calculate the driving probability corresponding to each driving segment based on a preset probability algorithm.

[0053] Optionally, the driving probability corresponding to each driving segment can be obtained according to the starting and ending nodes corresponding to each driving segment and the preset probability algorithm.

[0054] Step S208, determine each driving segment with a driving probability less than the preset probability threshold as an abnormal segment according to the preset probability threshold.

[0055] Step S210: Obtain the analysis result of the target vehicle according to the abnormal segments and driving segments in the driving trajectory.

[0056] In this embodiment, the number of abnormal segments in the driving trajectory can be compared with the number of driving segments to obtain the analysis result of whether the vehicle passing behavior of the target vehicle is abnormal.

[0057] Specifically, if the ratio of the number of abnormal segments to the total number of segments is higher than the preset ratio threshold, obtain the analysis result that the vehicle passing behavior is abnormal; otherwise, obtain the analysis result that the vehicle passing behavior is normal.

[0058] In summary, the vehicle passing behavior analysis method of this embodiment can accurately analyze whether there is an abnormality in the vehicle passing behavior of the target vehicle by calculating the driving probabilities of each driving segment of the target vehicle, thereby effectively reducing the occurrence of highway toll evasion.

[0059] Third Embodiment

[0060] Figure 3 The flow diagram of the vehicle passing behavior analysis method according to the third embodiment of the present application is shown. This embodiment is a specific implementation of the above step S206. As shown in the figure, this embodiment mainly includes the following steps:

[0061] Step S302: Query the segment probability matrix of the target road network according to the start and end nodes of the driving segment to obtain the first probability value of the driving segment.

[0062] In this embodiment, the segment probability matrix of the target road network stores the segment probability values corresponding to each driving segment formed by any two road network nodes in the target road network.

[0063] For example, the segment probability value P1 of the driving segment L1 formed by the road network node a and the road network node b (which can also be regarded as the transfer probability value P1 from the road network node a to the road network node b), and the segment probability value P2 of the driving segment L2 formed by the road network node a and the road network node c (which can also be regarded as the transfer probability value P2 from the road network node a to the road network node c). Accordingly, the first probability value of the driving segment can be directly queried from the segment probability matrix according to the start and end nodes of the driving segment.

[0064] It should be noted that the two road network nodes forming each driving segment in the segment probability matrix can be two adjacent road network nodes in the target road network, or can also be adjacent two road network nodes.

[0065] Step S304: Obtain the shortest path of the start and end nodes according to the start and end nodes of the driving segment and the target road network, and obtain the second probability value of the driving segment according to the shortest path.

[0066] In this embodiment, the shortest path between the start and end nodes can be calculated according to the start and end nodes of the driving segment and the mileage of each path corresponding to each path formed by any two adjacent road network nodes in the target road network, so as to determine the second probability value of the driving segment.

[0067] Step S306: Determine the larger value between the first probability value and the second probability value as the driving probability of the driving segment.

[0068] Specifically, the first probability value of the driving segment can be compared with the second probability value, and the larger value between the two can be determined as the driving probability corresponding to the driving segment.

[0069] In summary, by calculating the driving probability of the driving segment, the present application can accurately evaluate whether there is an abnormality in the driving trajectory of the target vehicle in the target road network, so as to effectively avoid the occurrence of high-speed toll evasion.

[0070] Fourth Embodiment

[0071] Figure 4 The flowchart of the vehicle passing behavior analysis method according to the fourth embodiment of the present application is shown. This embodiment can continue to execute following the above step S210, and is used to re-evaluate whether there is an abnormality in the vehicle passing behavior of the target vehicle by performing further segmentation on each driving segment when the analysis result of step 210 is that the vehicle passing behavior is abnormal.

[0072] It should be noted that when the analysis result of abnormal vehicle passing behavior is obtained in step S210, it can first be determined whether it is caused by reversing the start and end nodes of the driving segment identification, resulting in the opposite driving direction of the driving segment. If not, then continue to execute the following steps of the fourth embodiment to further evaluate whether the vehicle passing behavior is abnormal.

[0073] As shown in the figure, the vehicle passing behavior analysis method of this embodiment mainly includes the following:

[0074] Step S402: In response to the analysis result of abnormal vehicle passing behavior, determine the driving segments with the same start and end nodes as the segments to be cut according to the start and end nodes corresponding to each driving segment.

[0075] As described above, in step S204, the driving trajectory can be segmented based on two non-adjacent road network nodes in the target network to obtain each driving segment. In this case, if there are two or more driving segments with the same start and end nodes, it means that the current driving segment is unreasonably segmented and further refined segmentation needs to be performed.

[0076] Step S404: Split the segment to be cut according to each road network node in the target road network to obtain each first driving sub-segment.

[0077] Specifically, based on each road network node in the target road network, further splitting can be performed on the segment to be cut to obtain each first driving sub-segment.

[0078] In this embodiment, the starting and ending nodes (starting node and ending node) of each first driving sub-segment can be two adjacent road network nodes in the target road network, or two non-adjacent road network nodes.

[0079] Step S406: Based on a preset probability algorithm, calculate the driving probabilities corresponding to each first driving sub-segment, and update the analysis result of normal or abnormal vehicle passing behavior accordingly.

[0080] In this embodiment, the steps described in the above third embodiment can be used to calculate the driving probabilities corresponding to each first driving sub-segment, which will not be elaborated here.

[0081] Fifth Embodiment

[0082] Figure 5 The flowchart of the vehicle passing behavior analysis method according to the fifth embodiment of the present application is shown. This embodiment can continue to execute following the above step S210, and is used to re-evaluate whether there is an abnormality in the vehicle passing behavior of the target vehicle by performing further splitting on each driving segment when the analysis result of step 210 is abnormal vehicle passing behavior.

[0083] As shown in the figure, this embodiment mainly includes the following steps:

[0084] Step S502: Obtain the transaction medium of the target vehicle according to the entry data and exit data of the target vehicle.

[0085] In this embodiment, the transaction medium of the target vehicle refers to the medium used to pay the toll generated when the target vehicle passes through the target road network.

[0086] Optionally, the transaction types of the transaction medium of the target vehicle at least include one of cash, OBU, ETC, and CPC.

[0087] In this embodiment, the transaction medium of the target vehicle can be obtained according to the passing channel (such as the ETC channel or the manual toll collection channel) when the target vehicle enters the highway road network and the passing channel when the target vehicle exits the highway road network.

[0088] Step S504: If the transaction medium of the target vehicle contains at least two, split the driving segment based on the transaction medium to obtain each second driving sub-segment.

[0089] For example, if there are more than two transaction media for the target vehicle at the same time, it is possible that the target vehicle travels the same section of the target road network back and forth on the same day, and uses two transaction media to pay the toll. In this case, the driving section can be further segmented based on the transaction media to obtain each second driving sub-section.

[0090] Step S506: Based on a preset probability algorithm, calculate the driving probability corresponding to each second driving sub-section, and update the analysis result of normal or abnormal vehicle passing behavior accordingly.

[0091] In this embodiment, the steps described in the above third embodiment can be used to calculate the driving probability corresponding to each second driving sub-section, which will not be elaborated here.

[0092] Sixth Embodiment

[0093] Figure 6 Fig. shows a schematic flow chart of the vehicle passing behavior analysis method according to the fifth embodiment of the present application. This embodiment can combine the technical means of the above fourth and fifth embodiments to perform further fine segmentation on each driving section currently segmented for the target vehicle, so as to re-evaluate whether there is an abnormality in the vehicle passing behavior of the target vehicle.

[0094] As shown in the figure, this embodiment mainly includes the following steps:

[0095] Step S602: In response to the analysis result of abnormal vehicle passing behavior, based on the start and end nodes corresponding to each driving section, determine the driving sections with the same start and end nodes as the to-be-cut sections, and obtain the transaction media of the target vehicle according to the driving-in data and driving-out data of the target vehicle.

[0096] Step S604: If there are to-be-cut sections and at least two transaction media, divide the to-be-cut sections according to each road network node in the target road network to obtain each first driving sub-section, and / or divide the driving section based on the transaction media to obtain each second driving sub-section.

[0097] Step S606: Based on a preset probability algorithm, calculate the driving probability corresponding to each first driving sub-section and / or each second driving sub-section, and update the analysis result of normal or abnormal vehicle passing behavior accordingly.

[0098] In summary, in the fourth to sixth embodiments of this embodiment, when obtaining the analysis result of abnormal vehicle passing behavior, further fine cutting can be performed on each driving section according to the start and end nodes of each current driving section of the target vehicle and / or the transaction media of the target vehicle, and the vehicle passing behavior can be re-evaluated accordingly, thereby improving the accuracy of the analysis result.

[0099] Seventh Embodiment

[0100] Figure 7 Shows a schematic flow chart of the vehicle passing behavior analysis method according to the seventh embodiment of the present application. This embodiment can be executed following step S106. As shown in the figure, this embodiment mainly includes the following steps:

[0101] Step S702, in response to the analysis result of the abnormal vehicle passing behavior, predict the abnormal behavior category of the target vehicle according to the driving data.

[0102] Optionally, a trained classifier can be used to perform prediction according to the driving data and the preset category label to obtain the abnormal behavior category of the target vehicle.

[0103] In this embodiment, the abnormal behavior categories of the target vehicle may at least include: over-running, truck with passenger label, entry without exit, exit without entry, no entry and no exit, U-turn driving, shielding the transaction medium, large vehicle with small label, switching the transaction medium, incomplete deduction, one vehicle with multiple signatures, one vehicle with multiple cards.

[0104] In this embodiment, the confusion matrix can be used to train the classifier.

[0105] Specifically, the abnormal behavior category output by the classifier can be used as the prediction label, and the confusion matrix can be used to assist in calculating the precision value, recall value and F1 value according to the prediction label of the classifier and the manually marked true label, and accordingly train the classifier.

[0106] In summary, the vehicle passing behavior analysis method of this embodiment can predict the specific abnormal behavior category of the target vehicle with abnormal vehicle passing behavior to effectively inspect the abnormal data on the highway.

[0107] Eighth Embodiment

[0108] The eighth embodiment of the present application provides a computer storage medium, in which instructions for executing each step in the method described in any one of the first to seventh embodiments above are stored.

[0109] Ninth Embodiment

[0110] Figure 8 Shows a schematic architecture diagram of the vehicle passing behavior analysis device according to the ninth embodiment of the present application. As shown in the figure, the vehicle passing behavior analysis device 800 of this embodiment mainly includes a data acquisition module 802 and a behavior analysis module 804.

[0111] The data acquisition module 802 is used to continuously acquire the driving data of the target vehicle within a preset analysis time limit in response to the analysis instruction of the target vehicle.

[0112] Optionally, the driving data of the target vehicle includes at least one of: driving-in data of the target vehicle entering the target road network, driving-out data of the target vehicle leaving the target road network, and passing data of the target vehicle driving in the target road network.

[0113] Optionally, the analysis instruction includes automatic triggering that meets the triggering condition and / or manual triggering based on user input.

[0114] Optionally, the automatic triggering that meets the triggering condition includes: when the driving-out data of the target vehicle is acquired, triggering the analysis instruction; and / or dynamically updating the data reception time limit of the target vehicle according to the driving-in data or the passing data of the target vehicle, and if no new passing data or driving-out data is received within the data reception time limit, triggering the analysis instruction.

[0115] The behavior analysis module 804 is configured to analyze the vehicle passing behavior of the target vehicle according to the driving data, and obtain an analysis result indicating that the vehicle passing behavior is normal or abnormal.

[0116] Wherein, if the analysis result of the behavior analysis module 804 indicates that the vehicle passing behavior is abnormal, and the data acquisition module 802 acquires new driving data within the preset analysis time limit, the behavior analysis module 804 is provided to re-analyze the vehicle passing behavior of the target vehicle according to the new driving data, so as to update the analysis result.

[0117] Optionally, the behavior analysis module 804 further includes: obtaining the driving trajectory of the target vehicle in the target road network according to the driving-in data, the driving-out data, and the passing data of the target vehicle; segmenting the driving trajectory according to each road network node in the target road network to obtain each driving segment of the target vehicle; calculating each driving probability corresponding to each driving segment based on a preset probability algorithm; and obtaining the analysis result indicating that the vehicle passing behavior is normal or abnormal according to each driving probability corresponding to each driving segment.

[0118] Optionally, the behavior analysis module 804 further includes, for each driving segment, performing the following steps: querying the segment probability matrix of the target road network according to the start and end nodes of the driving segment to obtain a first probability value of the driving segment; obtaining the shortest path of the start and end nodes according to the start and end nodes of the driving segment and the target road network, and obtaining a second probability value of the driving segment according to the shortest path; and determining the larger one of the first probability value and the second probability value as the driving probability of the driving segment.

[0119] Optionally, each segment probability value corresponding to each driving segment formed by any two road network nodes in the target road network is stored in the segmented probability matrix.

[0120] Optionally, the behavior analysis module 804 further includes determining, according to a preset probability threshold, each driving segment with a driving probability less than the preset probability threshold as an abnormal segment; according to the number of abnormal segments and the number of driving segments of the driving segments, if the ratio of the number of abnormal segments to the number of driving segments is higher than a preset ratio threshold, obtaining the analysis result that the vehicle passing behavior is abnormal.

[0121] Optionally, the behavior analysis module 804 further includes, in response to the analysis result of abnormal vehicle passing behavior, determining, according to the starting and ending nodes corresponding to each driving segment, each driving segment with the same starting and ending nodes as a segment to be cut; cutting the segment to be cut according to each road network node in the target road network to obtain each first driving sub-segment; based on the preset probability algorithm, calculating each driving probability corresponding to each first driving sub-segment, and accordingly updating the analysis result of normal vehicle passing behavior or abnormal vehicle passing behavior.

[0122] Optionally, the behavior analysis module 804 further includes obtaining the transaction medium of the target vehicle according to the driving-in data and the driving-out data of the target vehicle; if the transaction medium of the target vehicle includes at least two, splitting the driving segment based on the transaction medium to obtain each second driving sub-segment; based on the preset probability algorithm, calculating each driving probability corresponding to each second driving sub-segment, and accordingly updating the analysis result of normal vehicle passing behavior or abnormal vehicle passing behavior.

[0123] Optionally, the transaction medium is used to pay the toll generated by the target vehicle passing through the target road network, where the transaction types of the transaction medium at least include one of cash, OBU, ETC, and CPC.

[0124] Optionally, the behavior analysis module 804 further includes obtaining the vehicle type specification, transaction fee, and driving mileage of the target vehicle according to the driving data of the target vehicle; obtaining the analysis result of normal vehicle passing behavior or abnormal vehicle passing behavior according to the vehicle type specification, the transaction fee, and the driving mileage.

[0125] Optionally, the behavior analysis module 804 further includes, in response to the analysis result of abnormal vehicle passing behavior, predicting the abnormal behavior category of the target vehicle according to the driving data.

[0126] Optionally, the behavior analysis module 804 further includes using a trained classifier to perform a prediction based on the driving data and a preset category label to obtain the abnormal behavior category of the target vehicle; wherein, the abnormal behavior category at least includes: overtime driving, passenger label for freight vehicle, entry without exit, exit without entry, no entry and no exit, U-turn driving, shielding the transaction medium, large vehicle with small label, switching the transaction medium, incomplete deduction, one vehicle with multiple signatures, one vehicle with multiple cards, etc.

[0127] In summary, the vehicle passing behavior analysis method, device and computer storage medium provided by the embodiments of the present application collect the driving data of the target vehicle, and accurately analyze whether there is an abnormality in the vehicle passing behavior of the target vehicle, so as to effectively reduce the occurrence of highway toll evasion.

[0128] In addition, by setting an analysis time limit in the present application, when the analysis result is that the vehicle passing behavior is abnormal and new driving data is obtained within the preset analysis time limit, the vehicle passing behavior of the target vehicle can be re-analyzed according to the new driving data, so as to avoid the problem that the analysis result is abnormal due to untimely transmission of driving data.

[0129] Furthermore, in the present application, by performing segmentation on the driving trajectory of the target vehicle according to the driving data of the target vehicle to obtain each driving segment of the target vehicle, and by calculating the driving probabilities corresponding to each driving segment, it is analyzed whether the vehicle passing behavior of the target vehicle is abnormal. In this way, the accuracy of the vehicle passing behavior analysis result can be improved.

[0130] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of the present application, and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for analyzing vehicle passing behaviors, characterized in that, Including: In response to an analysis instruction of a target vehicle, continuously obtain the driving data of the target vehicle within a preset analysis time limit, wherein the input order of the driving data does not depend on the chronological order; Analyze the vehicle passing behavior of the target vehicle according to the driving data, and obtain an analysis result of normal vehicle passing behavior or abnormal vehicle passing behavior; and If the analysis result is abnormal vehicle passing behavior and new driving data is obtained within the preset analysis time limit, re-analyze the vehicle passing behavior of the target vehicle according to the new driving data to update the analysis result; The analyzing the vehicle passing behavior of the target vehicle according to the driving data and obtaining an analysis result of normal vehicle passing behavior or abnormal vehicle passing behavior includes: obtaining the driving trajectory of the target vehicle in the target road network according to the driving data of the target vehicle; segmenting the driving trajectory according to each road network node in the target road network to obtain each driving segment of the target vehicle; obtaining each driving probability corresponding to each driving segment according to each start and end node corresponding to each driving segment and a preset probability algorithm, wherein the preset probability algorithm includes a segment probability matrix and a shortest path probability calculation rule, and each segment probability value corresponding to each driving segment formed by any two road network nodes in the target road network is stored in the segment probability matrix; each driving probability corresponding to each driving segment is the larger one among the probability values calculated according to the segment probability matrix and the shortest path probability calculation rule; obtaining the analysis result of normal vehicle passing behavior or abnormal vehicle passing behavior according to each driving probability corresponding to each driving segment.

2. The vehicle passing behavior analysis method according to claim 1, characterized in that The driving data of the target vehicle includes at least one of: the driving-in data of the target vehicle entering the target road network, the driving-out data of the target vehicle leaving the target road network, and the passing data of the target vehicle driving in the target road network.

3. The vehicle passing behavior analysis method according to claim 2, wherein The analysis instruction includes automatic triggering that meets the trigger condition and / or manual triggering based on user input.

4. The vehicle passing behavior analysis method according to claim 3, wherein The automatic triggering that meets the trigger condition includes: When the driving-out data of the target vehicle is obtained, trigger the analysis instruction; and / or Dynamically update the data reception time limit of the target vehicle according to the driving-in data or the passing data of the target vehicle. If no new passing data or driving-out data is received within the data reception time limit, trigger the analysis instruction.

5. The vehicle passing behavior analysis method according to claim 2, characterized in that, The analyzing the vehicle passing behavior of the target vehicle according to the driving data and obtaining an analysis result of normal vehicle passing behavior or abnormal vehicle passing behavior includes: Obtaining the driving trajectory of the target vehicle in the target road network according to the driving-in data, the driving-out data, and the passing data of the target vehicle; Segmenting the driving trajectory according to each road network node in the target road network to obtain each driving segment of the target vehicle; Calculating each driving probability corresponding to each driving segment based on a preset probability algorithm; Obtaining the analysis result of normal vehicle passing behavior or abnormal vehicle passing behavior according to each driving probability corresponding to each driving segment.

6. The vehicle passing behavior analysis method according to claim 5, wherein Calculating the driving probabilities corresponding to each of the driving segments based on the preset probability algorithm includes: For each of the driving segments, the following steps are performed: According to the start and end nodes of the driving segment, query the segment probability matrix of the target road network to obtain the first probability value of the driving segment; According to the start and end nodes of the driving segment and the target road network, obtain the shortest path of the start and end nodes, and according to the shortest path, obtain the second probability value of the driving segment; Determine the larger of the first probability value and the second probability value as the driving probability of the driving segment.

7. The vehicle passing behavior analysis method according to claim 6, wherein Obtaining the analysis result that the vehicle passing behavior is normal or the vehicle passing behavior is abnormal according to the driving probabilities corresponding to each of the driving segments includes: According to a preset probability threshold, determine the driving segments with driving probabilities less than the preset probability threshold as abnormal segments; According to the number of abnormal segments and the number of driving segments of the driving segments, if the ratio of the number of abnormal segments to the number of driving segments is higher than a preset ratio threshold, obtain the analysis result that the vehicle passing behavior is abnormal.

8. The vehicle passing behavior analysis method according to claim 7, wherein The method further includes: In response to the analysis result that the vehicle passing behavior is abnormal, according to the start and end nodes corresponding to each of the driving segments, determine the driving segments with the same start and end nodes as the segments to be cut; Cut the segments to be cut according to the road network nodes in the target road network to obtain each first driving sub-segment; Based on the preset probability algorithm, calculate the driving probabilities corresponding to each of the first driving sub-segments, and accordingly update the analysis result that the vehicle passing behavior is normal or the vehicle passing behavior is abnormal.

9. The vehicle passing behavior analysis method according to any one of claims 7 or 8, characterized in that The method further includes: According to the driving-in data and the driving-out data of the target vehicle, obtain the transaction medium of the target vehicle; If the transaction medium of the target vehicle includes at least two, based on the transaction medium, cut the driving segment to obtain each second driving sub-segment; Based on the preset probability algorithm, calculate the driving probabilities corresponding to each of the second driving sub-segments, and accordingly update the analysis result that the vehicle passing behavior is normal or the vehicle passing behavior is abnormal.

10. The vehicle passing behavior analysis method according to claim 9, wherein, The transaction medium is used to pay the toll generated by the target vehicle passing through the target road network, where the transaction types of the transaction medium include at least one of cash, OBU, ETC, and CPC.

11. The vehicle passing behavior analysis method according to claim 2, wherein, Analyzing the vehicle passing behavior of the target vehicle according to the driving data and obtaining the analysis result that the vehicle passing behavior is normal or the vehicle passing behavior is abnormal includes: According to the driving data of the target vehicle, obtain the vehicle type specification, transaction fee, and driving mileage of the target vehicle; According to the vehicle type specification, the transaction fee, and the driving mileage, obtain the analysis result that the vehicle passing behavior is normal or the vehicle passing behavior is abnormal.

12. The vehicle passing behavior analysis method according to claim 1, wherein The method further includes: In response to the analysis result that the vehicle passing behavior is abnormal, predict the abnormal behavior category of the target vehicle according to the driving data.

13. The vehicle passing behavior analysis method according to claim 12, wherein The method further includes: Using the trained classifier, perform prediction according to the driving data and the preset category labels to obtain the abnormal behavior category of the target vehicle; Among them, the abnormal behavior category at least includes: overtime driving, truck passenger label, entry without exit, exit without entry, no entry and no exit, U-turn driving, shielding the transaction medium, large vehicle with small label, switching the transaction medium, incomplete deduction, one vehicle with multiple signatures, one vehicle with multiple cards, etc.

14. A computer storage medium, characterized in that, The computer storage medium stores instructions for executing each step in the method described in any one of the above claims 1 to 13.

15. A vehicle passing behavior analysis device, characterized in that, It includes: A data acquisition module, configured to continuously acquire the driving data of the target vehicle within a preset analysis time limit in response to an analysis instruction of the target vehicle, wherein the input order of the driving data does not depend on the time sequence; A behavior analysis module, configured to analyze the vehicle passing behavior of the target vehicle according to the driving data to obtain an analysis result of normal vehicle passing behavior or abnormal vehicle passing behavior; Among them, if the analysis result of the behavior analysis module is abnormal vehicle passing behavior, and the data acquisition module acquires new driving data within the preset analysis time limit, the behavior analysis module is provided to re-analyze the vehicle passing behavior of the target vehicle according to the new driving data to update the analysis result; The analyzing the vehicle passing behavior of the target vehicle according to the driving data to obtain an analysis result of normal vehicle passing behavior or abnormal vehicle passing behavior includes: obtaining the driving trajectory of the target vehicle in the target road network according to the driving data of the target vehicle; segmenting the driving trajectory according to each road network node in the target road network to obtain each driving segment of the target vehicle; obtaining the driving probability corresponding to each driving segment according to the starting and ending nodes corresponding to each driving segment and a preset probability algorithm, wherein the preset probability algorithm includes a segment probability matrix and a shortest path probability calculation rule, and the segment probability matrix stores the segment probability values corresponding to each driving segment formed by any two road network nodes in the target road network; the driving probability corresponding to each driving segment is the larger one among the probability values calculated according to the segment probability matrix and the shortest path probability calculation rule; obtaining the analysis result of normal vehicle passing behavior or abnormal vehicle passing behavior according to the driving probabilities corresponding to each driving segment.

Images