CN112150810A

CN112150810A - Vehicle behavior management method, system, device and medium

Info

Publication number: CN112150810A
Application number: CN202011026385.0A
Authority: CN
Inventors: 周曦; 姚志强; 孙庆凯; 黄晓立; 何洪路; 张迪; 许梅芳; 赵科; 殷露曦; 杨华; 林尔彬; 林坤琳; 赵玉亮
Original assignee: Yuncong Technology Group Co Ltd
Current assignee: Yuncong Technology Group Co Ltd
Priority date: 2020-09-25
Filing date: 2020-09-25
Publication date: 2020-12-29

Abstract

The invention provides a vehicle behavior management method, a system, equipment and a medium, wherein traffic transaction data formed when one or more vehicles pass through a target area and vehicle image structured data related to the traffic transaction data are obtained; wherein the vehicle image structured data is obtained by analyzing an image of the vehicle passing through the target area; performing behavior management on the one or more vehicles based on the traffic transaction data and the vehicle image structured data. The mass data are analyzed by using a big data analysis technology, the driving condition of the vehicle on the highway is restored through the data, the abnormal driving behaviors in the vehicle can be mined, and the abnormal driving behaviors of one or more vehicles can be analyzed; and by storing the vehicle image structured data, the memory is saved more than the direct storage of the image, and the storage pressure is reduced compared with the prior art.

Description

Vehicle behavior management method, system, device and medium

Technical Field

The present invention relates to the field of image processing technologies, and in particular, to a method, a system, a device, and a medium for vehicle behavior management.

Background

There have been long-term abnormal driving behaviors of vehicles (such as rushing to the card, big and small traffic signs, running to buy short, etc.) on the highway, which can cause the highway to generate various strange charging data results. At present, in part of provinces, suspected fee evasion behavior data is obtained by a data analysis method, and certain results are obtained.

However, a large number of gantries are additionally arranged in a new toll collection system without a junction station, so that obvious changes are brought to the expressway. Firstly, the data volume of the highway is increased by multiple times, the national toll data can reach 7 hundred million days, and a big data analysis technology is needed to deeply mine the suspicious behavior expression in the data; secondly, a large number of gantries are arranged, so that the data collected at the front end of the highway can meet the requirement of fee evasion check business, and meanwhile, the storage pressure of mass visual data is brought; thirdly, when the vehicle escapes intentionally, the vehicle can be disguised intentionally by means of changing the license plate by masking and the like, so that the behavior data of the vehicle is difficult to track the real vehicle.

Disclosure of Invention

In view of the above-described drawbacks of the prior art, it is an object of the present invention to provide a vehicle behavior management method, system, device, and medium for solving the problems in the prior art.

To achieve the above and other related objects, the present invention provides a vehicle behavior management method, comprising the steps of:

acquiring traffic transaction data formed when one or more vehicles pass through a target area and vehicle image structured data associated with the traffic transaction data; wherein the vehicle image structured data is obtained by analyzing an image of the vehicle passing through the target area;

performing behavior management on the one or more vehicles based on the traffic transaction data and the vehicle image structured data.

Optionally, the behavior management of the one or more vehicles includes at least: analyzing whether abnormal driving behaviors exist in the one or more vehicles.

Optionally, the method further comprises:

restoring the driving track of the one or more vehicles according to the traffic transaction data and the vehicle image structured data;

and analyzing the recovered driving track according to a preset driving rule and/or a pre-constructed highway network model, and determining whether one or more vehicles have abnormal driving behaviors.

Optionally, the specific process of analyzing the recovered driving trajectory according to a pre-constructed highway network model and determining whether the one or more vehicles have abnormal driving behaviors includes:

carrying out unsupervised training on the existing road network data of the expressway through an artificial intelligence algorithm to construct an expressway road network model;

inputting the traffic transaction data and the vehicle image structured data into the highway road network model for simulation, and acquiring simulated driving tracks of one or more vehicles;

comparing the simulated driving track with the restored driving track; and analyzing whether the one or more vehicles have abnormal driving behaviors or not according to the comparison result.

Optionally, if there is an abnormal driving behavior of a certain vehicle, the driving behavior of the vehicle includes at least one of the following: the method comprises the steps that corresponding passing transaction data are not generated when a vehicle passes through a target area, mileage transaction amount when the vehicle generates the passing transaction data is not matched with corresponding driving mileage, the type of a toll vehicle is not consistent with a preset billing vehicle type when the vehicle generates the passing transaction data, the vehicle drives in two opposite directions in a single driving track, a plurality of vehicle-mounted electronic label device sets exist in the single driving track, and a plurality of composite pass cards exist in the single driving track.

Optionally, the specific process of acquiring the vehicle image structured data includes:

acquiring images of one or more vehicles passing through a target area, and recording the images as running images;

carrying out image recognition on the driving image, analyzing the driving image, and acquiring vehicle information of the one or more vehicles;

acquiring analyzed vehicle image structured data based on the vehicle information;

the vehicle information comprises license plate information, basic vehicle attributes, vehicle service attributes and vehicle identity marks.

Optionally, before restoring the driving track of the one or more vehicles, cleaning the traffic transaction data according to common knowledge and business rules of the highway; and separating, combining and cleaning the traffic transaction data and the license plate information according to the highway network rule.

Optionally, inputting a target image containing the target vehicle; searching out similar images meeting a preset similarity range from images of a vehicle passing through a target area through a visual intelligent engine, and displaying the similar images in an image list view; wherein the displayed information at least comprises image time and image similarity.

Optionally, inputting a target image containing the target vehicle; searching out similar images meeting a preset similarity range from images of a vehicle passing through a target area through a visual intelligent engine; and determining the running track of the target vehicle within the inquiry time range according to the similar image, and displaying the running track.

Optionally, the license plate information at least includes: license plate number and license plate color;

the vehicle basic attributes at least include: vehicle brand and vehicle type;

the vehicle service attributes include at least: and (4) hazardous chemicals.

Optionally, the method further comprises displaying the analysis result of the abnormal driving behavior of the vehicle, the recovered driving track and the simulated driving track by a visual chart;

wherein the form in which the chart is displayed comprises a dynamic form or a static form; the graph includes at least: map, thermodynamic diagram, pie chart, line chart, bar chart.

Optionally, audit analysis is performed according to the displayed chart, and audit information is fed back; and after the audit information is fed back, supervised training learning and optimization are carried out according to the abnormal driving behaviors and the simulated driving tracks of the vehicle.

Optionally, the target region comprises at least one of: toll stations at the entrance of the highway, toll stations at the exit of the highway and a portal frame between the entrance and the exit of the highway.

The invention also provides a vehicle behavior management system, comprising:

the data acquisition module is used for acquiring traffic transaction data formed when one or more vehicles pass through a target area and vehicle image structured data related to the traffic transaction data; wherein the vehicle image structured data is obtained by analyzing an image of the vehicle passing through the target area;

and the vehicle behavior management module is used for performing behavior management on the one or more vehicles according to the traffic transaction data and the vehicle image structured data.

Optionally, if the behavior management module of the vehicle manages the behavior of the one or more vehicles, analyzing whether the one or more vehicles have abnormal driving behavior; then there are:

Optionally, the specific process of analyzing the recovered driving trajectory by the vehicle behavior management module according to a pre-constructed highway network model and determining whether the one or more vehicles have abnormal driving behaviors includes:

Optionally, the system further comprises a first search module, configured to input a target image including a target vehicle; searching out similar images meeting a preset similarity range from images of a vehicle passing through a target area through a visual intelligent engine, and displaying the similar images in an image list view; wherein the displayed information at least comprises image time and image similarity.

Optionally, the system further comprises a second searching module, configured to input a target image including the target vehicle; searching out similar images meeting a preset similarity range from images of a vehicle passing through a target area through a visual intelligent engine; and determining the running track of the target vehicle within the inquiry time range according to the similar image, and displaying the running track.

The present invention also provides a vehicle behavior management apparatus including:

The present invention also provides an apparatus comprising:

one or more processors; and

one or more machine-readable media having instructions stored thereon that, when executed by the one or more processors, cause the apparatus to perform a method as in any one of the above.

The invention also provides one or more machine-readable media having instructions stored thereon, which when executed by one or more processors, cause an apparatus to perform the method as described in any one of the above.

As described above, the vehicle behavior management method, system, device, and medium provided by the present invention have the following beneficial effects: obtaining traffic transaction data formed when one or more vehicles pass through a target area and vehicle image structured data related to the traffic transaction data; the vehicle image structured data is obtained by analyzing an image when the vehicle passes through the target area; and performing behavior management on one or more vehicles based on the traffic transaction data and the vehicle image structured data. Wherein the behavior management of the one or more vehicles comprises at least: analyzing whether abnormal driving behaviors exist in the one or more vehicles. The invention analyzes mass data by using a big data analysis technology, restores the driving condition of the vehicle on the highway by data, can mine abnormal driving behaviors therein, and can analyze the abnormal driving behaviors of one or more vehicles. Meanwhile, a large number of vehicle driving images shot in a target area (such as a toll station at an entrance of an expressway, a toll station at an exit of the expressway, and a portal frame between the entrance and the exit of the expressway) are analyzed into vehicle image structured data through an AI (AI) vision technology, and the memory is saved by storing the vehicle image structured data rather than directly storing the images, so that the storage pressure is reduced. The method can quickly and effectively apply valuable vehicle information in the image to form vehicle identity information (ReiD) independent of license plate recognition; after AI machine learning is introduced, new abnormal behaviors can be found in an unsupervised training mode, and a good closed loop of human-computer cooperation and mutual growth check escape business work can be formed in a supervised training mode.

Drawings

FIG. 1 is a schematic flow chart diagram illustrating a vehicle behavior management method according to an exemplary embodiment;

FIG. 2 is a schematic flow chart diagram of a vehicle behavior management method according to another embodiment;

FIG. 3 is a road map of a driving track of a vehicle according to an embodiment;

FIG. 4 is a schematic view of a portal and toll booth associated with the vehicle travel track route of FIG. 3;

fig. 5 is a schematic hardware configuration diagram of a vehicle behavior analysis system according to an embodiment;

fig. 6 is a schematic hardware structure diagram of a terminal device according to an embodiment;

fig. 7 is a schematic diagram of a hardware structure of a terminal device according to another embodiment.

Description of the element reference numerals

M10 data acquisition module

M20 vehicle behavior management module

1100 input device

1101 first processor

1102 output device

1103 first memory

1104 communication bus

1200 processing assembly

1201 second processor

1202 second memory

1203 communication assembly

1204 Power supply Assembly

1205 multimedia assembly

1206 Audio component

1207 input/output interface

1208 sensor assembly

Detailed Description

The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.

It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the components related to the present invention are only shown in the drawings rather than drawn according to the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated.

Referring to fig. 1, the present invention provides a vehicle behavior management method, including the following steps:

s100, acquiring traffic transaction data formed when one or more vehicles pass through a target area and vehicle image structured data related to the traffic transaction data; the vehicle image structured data is obtained by analyzing an image when the vehicle passes through the target area;

and S200, performing behavior management on one or more vehicles based on the traffic transaction data and the vehicle image structured data.

The method analyzes mass data by using a big data analysis technology, recovers the driving condition of the vehicle on the highway through the data, and excavates abnormal driving behaviors in the vehicle; the method solves the problem that the mass data of the highway generates higher value application outside the charging service, so that the user can know the driving condition of the highway vehicle and gradually excavate fee evasion behaviors and vehicles thereof. Meanwhile, a large number of vehicle driving images shot in the target area are analyzed into vehicle image structured data through an AI vision technology, and the memory is saved by storing the vehicle image structured data compared with directly storing the images, so that the storage pressure is reduced. The problem that massive images bring storage pressure and are not beneficial to application is solved, the structured data after image analysis meets the requirement of highway business, the size is proper, storage is facilitated, and the structured mode is more suitable for big data analysis and operation. Meanwhile, the image analysis supports the optimization processing of different angles and different light ray images, so that the vehicle information in the low-quality images is finally obtained. Moreover, the image analysis enables a user to obtain a vehicle identity (ReiD) independent of a license plate, and the identity has certain confidentiality through vector storage; vector search is supported, image information can be quickly searched in service application, and the effect of inquiring billions of image data in seconds is achieved.

In an exemplary embodiment, the behavior management of one or more vehicles includes at least: and analyzing whether one or more vehicles have abnormal driving behaviors. Specifically, the driving tracks of one or more vehicles are restored according to the traffic transaction data and the vehicle image structured data; and analyzing the recovered driving track according to a preset driving rule and/or a pre-constructed highway network model to determine whether one or more vehicles have abnormal driving behaviors. In the embodiment of the present application, the preset driving rule is a driving rule empirically determined by the fact that fee evasion is known on the expressway. According to the embodiment, the recovered driving track can be analyzed empirically through the known fee evasion facts of the expressway, the fee evasion behavior track with high suspicion is screened out, and whether abnormal driving behaviors exist in one or more vehicles is determined. In the embodiment of the present application, the specific process of analyzing the recovered driving trajectory according to the pre-constructed highway network model and determining whether one or more vehicles have abnormal driving behaviors includes: carrying out unsupervised training on the existing road network data of the expressway through an artificial intelligence algorithm to construct an expressway road network model; inputting the traffic transaction data and the vehicle image structured data into a highway network model for simulation, and acquiring the simulated driving track of one or more vehicles; comparing the simulated driving track with the restored driving track; and analyzing whether one or more vehicles have abnormal driving behaviors or not according to the comparison result.

According to the above description, the present embodiment uses the existing highway network parameters, including but not limited to: the method comprises the steps of selecting a proper AI algorithm to construct an expressway network model, wherein the method comprises the following steps of toll road (Tollroad), toll road Section (Section), toll inter (TollInterval), toll station (TollStation), toll portal (TollGantry), network node coding information (NodeCode), adjacent node relation information (NodeRelation) and the like. The model is used as a road network basic model, can effectively judge interference information such as reverse cross-labeling and the like, is combined with highway audit service elements, and finally forms a mature model which can be more suitable for audit escape service. The method applies the AI model to truly reflect the relationship characteristics of the highway network, and simultaneously converts the complex road network into a mathematical model more suitable for calculation by means of AI algorithm characteristic engineering processing and the like, thereby achieving the effects of reducing the retrieval complexity and improving the performance. The traffic transaction data and the vehicle image structured data collected from the expressway are input into the road network model, and the traffic transaction data and the vehicle image structured data not only comprise information such as traffic media and vehicle identification license plates adopted by the expressway, but also comprise information such as vehicle identification marks (ReiD) obtained by analysis of the AI visual perception platform, so that the driving track of the vehicle on the expressway is conveniently restored. Under this business scene, the true orbit of traveling of vehicle decomposes and can obtain OBU information track point, ETC card information track point, CPC card information track point, discernment license plate track point, no tablet vehicle track point, vehicle identity sign track point and so on. The most real driving track of the vehicle can be restored by combining an AI algorithm through reasonable driving rule sets in the road network, track point combined path collision analysis and the like. Meanwhile, the AI can provide a high-confidence simulation driving track for the missing part in the track according to the road network characteristics.

In an exemplary embodiment, the specific process of acquiring the vehicle image structured data includes: acquiring images of one or more vehicles passing through a target area, and recording the images as running images; carrying out image recognition on the driving image, analyzing the driving image and acquiring vehicle information of one or more vehicles; acquiring the analyzed vehicle image structured data based on the vehicle information; the vehicle information comprises license plate information, basic vehicle attributes, vehicle service attributes and vehicle identity identifiers (ReiD). In the embodiment of the present application, the license plate information at least includes: license plate number and license plate color; the vehicle basic attributes at least include: vehicle brand and vehicle type; the vehicle service attributes include at least: and (4) hazardous chemicals.

According to the above description, the traffic transaction data acquired by the method can be transmitted to the data processing center in real time for storage, and meanwhile, the vehicle image structured data, the preset driving rule and the highway network model can also be stored in the data processing center. As an example, traffic transaction data formed when one or more vehicles pass through the target area is obtained and transmitted to the data processing center in real time; pushing the image code associated with the traffic transaction data to a branch node connected with the data processing center, and acquiring a driving image corresponding to the image code from the target area through the corresponding branch node; analyzing the driving image to obtain vehicle image structured data related to the traffic transaction data; performing behavior management on one or more vehicles based on the traffic transaction data and the vehicle image structured data; for example, whether one or more vehicles have abnormal driving behavior.

According to the above description, in some exemplary embodiments, if there is an abnormal traveling behavior of a certain vehicle, the traveling behavior of the certain vehicle includes at least one of: the method comprises the steps that corresponding passing transaction data are not generated when a vehicle passes through a target area, mileage transaction amount when the vehicle generates the passing transaction data is not matched with corresponding driving mileage, the type of a toll vehicle is not consistent with a preset billing vehicle type when the vehicle generates the passing transaction data, the vehicle drives in two opposite directions in a single driving track, a plurality of vehicle-mounted electronic label device sets exist in the single driving track, and a plurality of composite pass cards exist in the single driving track.

Specifically, as an example, if the vehicle does not generate corresponding transit transaction data as it passes through the target area, it may be that the vehicle is screened from the transit medium. For example, when the vehicle approaches the portal frame, the portal frame generates corresponding passing transaction data or image data according to the vehicle and a passing medium held by the vehicle; when the situation that continuous non-portal data exists on a vehicle passing path, the vehicle has malicious illegal fee evasion behaviors in modes of interfering an ETC (Electronic Toll Collection, non-cash card) label, shielding a highway composite passing card (CPC card for short) and the like at a high probability.

As another example, if the mileage transaction amount at which the vehicle generates the pass transaction data does not match the corresponding mileage, it may be that the vehicle is running long and buying short. For example, when the vehicle has an export transaction amount in a province, the transaction amount which the vehicle should receive is estimated through the recovered vehicle passing route, if the amount is obviously different from the export transaction amount of the vehicle, the passing is suspected to have a long run and a short run, and the passing is thrown as an abnormal condition so as to further analyze the possible fee evasion condition.

As another example, if the model of the toll vehicle is not consistent with the preset toll vehicle model when the vehicle generates the traffic transaction data, the vehicle may be a truck passenger mark or a large car small mark. For example, the actual vehicle type of the vehicle is screened by using the vehicle image structured data, the vehicle type is compared with the highway charging vehicle type, and the vehicle type is output as an abnormal condition when the charging vehicle type is found to be smaller than the actual vehicle type of the vehicle, so that the highway toll loss caused in the vehicle passing process can be timely paid. If the vehicle actually adopts ETC to pass, the vehicle can be required to go to a distributor in time to replace a vehicle-mounted electronic tag Unit (OBU) suit according with the vehicle type through the abnormal analysis of the large vehicle logo and the small vehicle logo. The abnormal behaviors comprise a goods vehicle passenger mark and a big vehicle small mark. The type of the freight car adopts the type of the passenger car to charge; the truck vehicle adopts a passenger car OBU to run on a highway. According to the current highway operation rule, the freight charge of the same grade is increased by multiplying the passenger charge amount by a coefficient, so the type inevitably causes toll loss. And the AI vision discriminates that the obtained vehicle is actually a truck type, and the vehicle type charged at the outlet of the obtained vehicle is a passenger car, so that the obtained vehicle is obtained by comparison. The large and small vehicle signs, the freight cars or the passenger cars adopt the charging and fee deduction of the vehicle type which is smaller than the actual vehicle type at the exit, including that the adopted OBU vehicle type is smaller than the actual vehicle type, or the adopted CPC card writing vehicle type is smaller than the actual vehicle type. According to the current highway operation rule, the larger the vehicle type, the higher the toll, so the type inevitably causes toll loss. And the AI vision discriminates and obtains the actual vehicle type of the truck or the passenger car, and compares the actual vehicle type with the vehicle type of the outlet charging of the truck or the passenger car to obtain the actual vehicle type.

As another example, if the vehicle is traveling in two opposite directions in a single travel trajectory, there may be a u-turn travel of the vehicle. For example, the vehicle exhibits reverse direction travel with no reachable tip-off position in a single travel trajectory. The abnormal behaviors may cause the suspicion of fee evasion of passing through or circulating running in the network. AI discrimination: the vehicle appears in the opposite direction on a running track without a head-up position within a short time or appears in the opposite direction after a long time and still adopts the same (secondary) passing medium.

As another example, if a vehicle has multiple sets of in-vehicle electronic label devices in a single travel trajectory, the vehicle may have one lot of labels. For example, a vehicle may have multiple OBU packages traveling on a highway at the same time. This type may raise a suspicion of fee evasion to switch or counterfeit ETC. AI discrimination: there are multiple OBU packages in a single travel trajectory.

As another example, if a vehicle has multiple composite transit cards in a single travel trajectory, the vehicle may have one or more cards. For example, a vehicle is traveling on a highway with multiple CPC cards simultaneously. This type may trigger either CPC card loss or violation of toll suspicion. AI discrimination: there are multiple CPC cards in a single travel trajectory.

In the embodiment of the present application, there may be other suspected evasive exception types according to the business experience in the past or in the future without being limited to the above listed exception types.

According to the above description, in an exemplary embodiment, before restoring the driving trajectory of one or more vehicles, the method further includes cleaning the traffic transaction data according to the common knowledge and the business rules of the highway; and separating, combining and cleaning the traffic transaction data and the license plate information according to the highway network rule. According to the embodiment of the application, the obtained traffic transaction original data is cleaned through the common sense of the expressway and the business rules, so that the interference data caused by card issuing errors and the like can be eliminated. And the vehicle running condition of the expressway is conveniently recovered by separating, combining and cleaning the traffic transaction data and the license plate information obtained by image analysis through the expressway network rules. In an exemplary embodiment, searching for a car with a map and searching for a track with a map may also be performed. Searching vehicles by images, which comprises inputting target images containing target vehicles to a data processing center; searching out similar images meeting a preset similarity range from driving images through a visual AI intelligent engine in a data processing center, and displaying the similar images in an image list view; wherein the displayed information at least comprises image time and image similarity. Searching the track by the image, wherein the step of inputting a target image containing a target vehicle to the data processing center is included; searching out similar images meeting a preset similarity range from driving images through a visual intelligent engine in a data processing center; and determining the running track of the target vehicle within the query time range according to the similar images, and displaying the running track. The embodiment of the application solves the problem that vehicles are disguised in a fake plate and fake plate mode, and has a vehicle identification mode which is independent of license plate identification and has high accuracy, so that users can use the mode to perform business applications such as vehicle searching and track searching by images. In the embodiment of the present application, when performing a graph search, the search is performed in a one-to-many manner, that is, 1: n searching; and searching a result which accords with a certain similarity range with the input target vehicle image in the target library by using a vector search mode through inputting the target vehicle image. Vehicle 1 in the embodiment of the present application: after the N search engine supports vehicle image analysis, billions of image vector value matching searches can be performed. In the conventional technology, aiming at massive data characteristics, approximate search is adopted to meet the requirement on speed, so that certain precision loss is caused, and the large-scale vector search is adopted to solve the problem of approximate neighbor search in the embodiment of the application; according to different scenes, different index searching modes are constructed, and the searching speed can be improved by several times or even dozens of times. In the embodiment of the application, the vector technology engine is applied to a vehicle image retrieval application scene, and comprises brute force search, inverted index, graph-based search and the like aiming at hash features. After the index is constructed, the method has higher retrieval speed and higher recall rate compared with the traditional mode, so that retrieval of various characteristics can be supported aiming at rich service scenes. Meanwhile, the vehicle ReiD can extract features for each vehicle, and vehicle ReiD and cross-border tracking are realized according to the features. The technology can be used as an important supplement of vehicle identity recognition, even if the vehicle license plate is removed, the vehicle can still be recognized according to algorithm application such as visual layering in the snapshot image, the recognition process does not need to depend on the vehicle license plate, recognition can be realized even if the vehicle is in a fake license plate state, and the vehicle running track can be continuously tracked, so that the behavior intention of a vehicle driver can be more completely analyzed. The analysis engine can further convert the extracted structural information into information and clues required in actual combat application in a real scene, so that the conversion of visual data into effective data is realized, and support is provided for super files, multi-dimensional retrieval and fine matching; vehicle retrieval, full/incremental clustering, and various structured retrieval can be achieved.

In some exemplary embodiments, the method further comprises displaying the analysis result of the abnormal driving behavior of the vehicle, the driving track recovered according to the traffic transaction data and the vehicle image structured data and the simulated driving track simulated by the highway network model in a visualized chart; wherein the form of displaying the diagram comprises a dynamic form or a static form; the graph includes at least: map, thermodynamic diagram, pie chart, line chart, bar chart. By way of example, the embodiment of the application can show the behavior track of the vehicle on the expressway in a chart mode and the like, so that the user can intuitively understand the behavior track and possible problems, and the user is supported to record the analysis result of the behavior track. For example, displaying the possible events of the highway attention area in a special highlight form; the area of interest includes toll booths and gantries, and the event of interest includes known anomalous behavior.

In the embodiment of the application, track query can be performed on the chart display interface; for example, providing conditions such as license plate, time, position and the like, searching all running tracks of the vehicle obtained by recovering the AI model simulation platform; and the specific details of a single track can be checked, and the comparison and display of multiple selected tracks can be supported. Wherein, the specific details of the single track are checked, and the specific details comprise: the map and animation display mode supports track decomposition and combined display; the detailed track point data display of the single track comprises information such as geographic positions, used passing media, related mileage/time/average speed and the like; the method comprises the following steps of (1) displaying vehicle file profiles and vehicle historical tracks; and (4) graphically displaying the single track, and comparing and displaying selected track points. Selecting a plurality of tracks for comparison and display comprises the following steps: the same vehicle can run for many times, and different vehicle running tracks can be supported; the map visually and visually shows the track comparison effect, and the page shows the data set of each track point, so that the similarity and difference between tracks can be quickly compared.

According to the above description, in some exemplary embodiments, the method further includes performing audit analysis according to the displayed chart, and feeding back audit information; and after the audit information is fed back, supervised training learning and optimization are carried out according to the abnormal driving behaviors and the simulated driving tracks. In the embodiment of the application, the input becomes a training example based on the information result explicitly fed back by the user. The AI, after obtaining targeted supervised learning, generates a supervised learning algorithm that analyzes the training data and makes its inference to produce an optimal solution, i.e., to allow the algorithm to correctly determine the class label without the label being visible. By receiving result data formed after the abnormal behaviors are analyzed by a user, training and learning are carried out on the result data by using the technologies such as AI machine learning and the like, an AI algorithm model is formed and is put into the abnormal behavior analysis and AI identification abnormality again for application, so that the abnormal behaviors with higher accuracy and (reformed) abnormal types and abnormal probabilities are produced. Based on the feedback information formed after manual checking, new available data analysis logic can be further deduced, and the system or the model can be optimized through modifying program implementation. The training learning method in the application can change the process into an artificial intelligence learning process automatically realized by a system or a model.

In accordance with the foregoing, in some exemplary embodiments, the target area includes at least one of: toll stations at the entrance of the highway, toll stations at the exit of the highway and a portal frame between the entrance and the exit of the highway.

According to the above description, in the embodiment of the provincial highway AI audit, as shown in fig. 2, the vehicles pass through the toll gate/portal, form the traffic transaction data, and upload the total amount of the formed traffic transaction data to the provincial center. The provincial center in this embodiment is a data processing center in some embodiments. And after the provincial center acquires the passing transaction data, pushing the passing transaction data to the AI audit for data butt joint. The AI audit sends image codes (including request image analysis) associated with the traffic transaction data to corresponding sub-nodes, the sub-nodes acquire driving images from corresponding toll stations or portals, the sub-nodes analyze the driving images, the analyzed and acquired vehicle image structured data are uploaded to a provincial center, and the provincial center stores the vehicle image structured data. Then, carrying out big data analysis on the traffic transaction data and the vehicle image structured data by AI audit to determine whether the vehicle has abnormal driving behaviors; and pushing the determined result to a display interface of the service foreground for display. When the inquiry user operates and triggers to view the image in the service foreground, the provincial center directly sends a corresponding viewing request to the toll station or the portal to obtain and display the corresponding driving image.

In one embodiment, as shown in fig. 3 and 4, when the AI audit is performed on the vehicle corresponding to the license plate of the Mina, it is found that the vehicle turns around in 26 days (2020 years and the same in the following years) in 7 months by using the ETC.

Analyzing the characteristics of the vehicle running path:

path one: 7, month, 26, day 03: 04 driving in a rabbeted pit, and 4: 01 disappears after passing through a Xiancun-southern Xiangshan portal frame;

and a second route: 5: 40 east rows appeared on the fairy village-stone beach portal, 6: 47 disappear after passing through the guan portal of Wu Pong-Hui Wan.

In this process, the vehicle pass is: 000000123456789456123456789456123450. the vehicle's pass is the image code associated with the transit transaction data in some embodiments.

The entries for the past are: 7, month 25, day 05: 47 enters the Huakuai three-phase synchronization station, 7 months, 26 days and 8 days: 46 on Shanshu high-speed boarding post-Shaxi portal.

And (3) man-machine cooperation verification:

1) and the OBU of the vehicle corresponding to the license plate A of the Min passes through the portal frame without charging records.

2) And the vehicles corresponding to the license plate A of the Min have no fee deduction record at the exit of the expressway.

3) And no portal frame picture record is carried out on the vehicle corresponding to the license plate of Min A.

4) And comparing the portal snapshot time with the OBU identification time, and applying the MinA license plate to the OBU corresponding to the vehicle corresponding to the Guangdong A license plate.

Other evidence collection: the vehicle searching with the picture finds that the vehicle corresponding to the YueA license plate has another license plate YueS.

And (3) calculating fee evasion amount: the vehicles corresponding to the Guangdong A license plate have 8 times of records of card input and output within five days, and the total fee of escape is 1800 Yuan. The vehicles corresponding to the other set of Guangdong license plates have 2 times of records of card admission and non-discharge within five days, and the total fee of escape is 340 yuan.

In conclusion, the highway has the nationwide networking charging parameters which are universal in the customized industry and the charging modules of each province carry out path reduction and charging on the vehicle passing condition, and the method simplifies the processing mode through data analysis, improves the data processing speed and can meet the requirement of the audit service. The method analyzes mass data by using a big data analysis technology, recovers the driving condition of the vehicle on the highway through the data, and excavates abnormal driving behaviors in the vehicle; meanwhile, a large number of vehicle driving images shot in the target area are analyzed into vehicle image structured data through an AI vision technology, and the memory is saved by storing the vehicle image structured data compared with directly storing the images, so that the storage pressure is reduced. The method can quickly and effectively apply valuable vehicle information in the image to form vehicle identity information ReID independent of license plate recognition; after AI machine learning is introduced, new abnormal behaviors can be found in an unsupervised training mode, and a good closed loop of human-computer cooperation and mutual growth check escape business work can be formed in a supervised training mode. The method can quickly and effectively process million-level data, and suspect behavior data with high accuracy and service value is screened out through data analysis; the data percentage may be 0.37% (the data still belongs to the suspect data, and the fee evasion fact can be confirmed by manual confirmation, and the data can obtain the empirical value for confirming the fee evasion percentage after the customer applies for a certain time). And the method can fuse the highway data and the artificial feedback information to form a self-learning growing closed-loop service flow. Meanwhile, highway charging data are converted into track information records with strong readability and reflecting vehicle behavior characteristics, track data meeting the requirement of the audit service are output through analysis and screening, a feedback result is formed after manual checking, and the result is used as input to conduct training optimization. In the image analysis stage, the method can manually mark the features of the vehicles in the image (such as license plates, head annual inspection labels and the like), and search a result image similar to the features of the target vehicle through feature matching.

As shown in fig. 5, the present invention also provides a vehicle behavior management system, including:

the data acquisition module M10 is used for acquiring traffic transaction data formed when one or more vehicles pass through the target area and vehicle image structured data related to the traffic transaction data; the vehicle image structured data is obtained by analyzing an image when the vehicle passes through the target area;

and the vehicle behavior management module M20 is used for performing behavior management on one or more vehicles according to the traffic transaction data and the vehicle image structured data.

The system analyzes mass data by using a big data analysis technology, restores the driving condition of a vehicle on the highway by data and excavates abnormal driving behaviors in the vehicle; the method solves the problem that the mass data of the highway generates higher value application outside the charging service, so that the user can know the driving condition of the highway vehicle and gradually excavate fee evasion behaviors and vehicles thereof. Meanwhile, the system analyzes a large number of vehicle running images shot in the target area into vehicle image structured data through an AI vision technology, and saves more memory than directly storing the images by storing the vehicle image structured data, thereby reducing the storage pressure. The problem that massive images bring storage pressure and are not beneficial to application is solved, the structured data after image analysis meets the requirement of highway business, the size is proper, storage is facilitated, and the structured mode is more suitable for big data analysis and operation. Meanwhile, the image analysis supports the optimization processing of different angles and different light ray images, so that the vehicle information in the low-quality images is finally obtained. Moreover, the image analysis enables a user to obtain a vehicle identity (ReiD) independent of a license plate, and the identity has certain confidentiality through vector storage; vector search is supported, image information can be quickly searched in service application, and the effect of inquiring billions of image data in seconds is achieved.

In an exemplary embodiment, the behavior management of one or more vehicles by the vehicle behavior management module M20 includes at least: and analyzing whether one or more vehicles have abnormal driving behaviors. Specifically, the driving tracks of one or more vehicles are restored according to the traffic transaction data and the vehicle image structured data; and analyzing the recovered driving track according to a preset driving rule and/or a pre-constructed highway network model to determine whether one or more vehicles have abnormal driving behaviors. In the embodiment of the present application, the preset driving rule is a driving rule empirically determined by the fact that fee evasion is known on the expressway. According to the embodiment, the recovered driving track can be analyzed empirically through the known fee evasion facts of the expressway, the fee evasion behavior track with high suspicion is screened out, and whether abnormal driving behaviors exist in one or more vehicles is determined. In the embodiment of the present application, the specific process of analyzing the recovered driving trajectory according to the pre-constructed highway network model and determining whether one or more vehicles have abnormal driving behaviors includes: carrying out unsupervised training on the existing road network data of the expressway through an artificial intelligence algorithm to construct an expressway road network model; inputting the traffic transaction data and the vehicle image structured data into a highway network model for simulation, and acquiring the simulated driving track of one or more vehicles; comparing the simulated driving track with the restored driving track; and analyzing whether one or more vehicles have abnormal driving behaviors or not according to the comparison result.

According to the above description, the present embodiment uses the existing highway network parameters, including but not limited to: the method comprises the steps of selecting a proper AI algorithm to construct an expressway network model, wherein the method comprises the following steps of toll road (Tollroad), toll road Section (Section), toll inter (TollInterval), toll station (TollStation), toll portal (TollGantry), network node coding information (NodeCode), adjacent node relation information (NodeRelation) and the like. The model is used as a road network basic model, can effectively judge interference information such as reverse cross-labeling and the like, is combined with highway audit service elements, and finally forms a mature model which can be more suitable for audit escape service. The system can truly reflect the relationship characteristics of the highway network by applying the AI construction model, and simultaneously converts the complex road network into a mathematical model more suitable for calculation by means of AI algorithm characteristic engineering processing and the like, thereby achieving the effects of reducing the retrieval complexity and improving the performance. The traffic transaction data and the vehicle image structured data collected from the expressway are input into the road network model, and the traffic transaction data and the vehicle image structured data not only comprise information such as traffic media and vehicle identification license plates adopted by the expressway, but also comprise information such as vehicle identification marks (ReiD) obtained by analysis of the AI visual perception platform, so that the driving track of the vehicle on the expressway is conveniently restored. Under this business scene, the true orbit of traveling of vehicle decomposes and can obtain OBU information track point, ETC card information track point, CPC card information track point, discernment license plate track point, no tablet vehicle track point, vehicle identity sign track point and so on. The most real driving track of the vehicle can be restored by combining an AI algorithm through reasonable driving rule sets in the road network, track point combined path collision analysis and the like. Meanwhile, the AI can provide a high-confidence simulation driving track for the missing part in the track according to the road network characteristics.

According to the above description, the system can transmit the traffic transaction data to the data processing center in real time for storage after acquiring the traffic transaction data, and meanwhile, the vehicle image structured data, the preset driving rule and the highway network model are also stored in the data processing center. As an example, traffic transaction data formed when one or more vehicles pass through the target area is obtained and transmitted to the data processing center in real time; pushing the image code associated with the traffic transaction data to a branch node connected with the data processing center, and acquiring a driving image corresponding to the image code from the target area through the corresponding branch node; analyzing the driving image to obtain vehicle image structured data related to the traffic transaction data; performing behavior management on one or more vehicles based on the traffic transaction data and the vehicle image structured data; for example, whether one or more vehicles have abnormal driving behavior.

In an exemplary embodiment, the system further comprises a data cleaning module, which is used for cleaning the traffic transaction data according to the common knowledge and the business rules of the highway before restoring the driving track of one or more vehicles; and separating, combining and cleaning the traffic transaction data and the license plate information according to the highway network rule. According to the embodiment of the application, the obtained traffic transaction original data is cleaned through the common sense of the expressway and the business rules, so that the interference data caused by card issuing errors and the like can be eliminated. And the vehicle running condition of the expressway is conveniently recovered by separating, combining and cleaning the traffic transaction data and the license plate information obtained by image analysis through the expressway network rules.

In an exemplary embodiment, the vehicle searching system further comprises a first searching module, and the first searching module is used for searching vehicles according to the images. The method comprises the following steps: inputting a target image containing a target vehicle to a data processing center; searching out similar images meeting a preset similarity range from driving images through a visual intelligent engine in a data processing center, and displaying the similar images in an image list view; wherein the displayed information at least comprises image time and image similarity. And the second searching module is used for searching tracks by using the images. The method comprises the following steps: inputting a target image containing a target vehicle to a data processing center; searching out similar images meeting a preset similarity range from driving images through a visual intelligent engine in a data processing center; and determining the running track of the target vehicle within the inquiry time range according to the similar images, and displaying the running track. The embodiment of the application solves the problem that vehicles are disguised in a fake plate and fake plate mode, and has a vehicle identification mode which is independent of license plate identification and has high accuracy, so that users can use the mode to perform business applications such as vehicle searching and track searching by images. In the embodiment of the present application, when performing a graph search, the search is performed in a one-to-many manner, that is, 1: n searching; and searching a result which accords with a certain similarity range with the input target vehicle image in the target library by using a vector search mode through inputting the target vehicle image. Vehicle 1 in the embodiment of the present application: after the N search engine supports vehicle image analysis, billions of image vector value matching searches can be performed. In the conventional technology, aiming at massive data characteristics, approximate search is adopted to meet the requirement on speed, so that certain precision loss is caused, and the large-scale vector search is adopted to solve the problem of approximate neighbor search in the embodiment of the application; according to different scenes, different index searching modes are constructed, and the searching speed can be improved by several times or even dozens of times. In the embodiment of the application, the vector technology engine is applied to a vehicle image retrieval application scene, and comprises brute force search, inverted index, graph-based search and the like aiming at hash features. After the index is constructed, the method has higher retrieval speed and higher recall rate compared with the traditional mode, so that retrieval of various characteristics can be supported aiming at rich service scenes. Meanwhile, the vehicle ReiD can extract features for each vehicle, and vehicle ReiD and cross-border tracking are realized according to the features. The technology can be used as an important supplement of vehicle identity recognition, even if the vehicle license plate is removed, the vehicle can still be recognized according to algorithm application such as visual layering in the snapshot image, the recognition process does not need to depend on the vehicle license plate, recognition can be realized even if the vehicle is in a fake license plate state, and the vehicle running track can be continuously tracked, so that the behavior intention of a vehicle driver can be more completely analyzed. The analysis engine can further convert the extracted structural information into information and clues required in actual combat application in a real scene, so that the conversion of visual data into effective data is realized, and support is provided for super files, multi-dimensional retrieval and fine matching; vehicle retrieval, full/incremental clustering, and various structured retrieval can be achieved.

In an exemplary embodiment, the system further comprises a display module, which is used for displaying the analysis result of the abnormal driving behavior of the vehicle, the driving track recovered according to the traffic transaction data and the vehicle image structured data and the simulated driving track simulated by the highway network model by using a visual chart; wherein the form of displaying the diagram comprises a dynamic form or a static form; the graph includes at least: map, thermodynamic diagram, pie chart, line chart, bar chart. By way of example, the embodiment of the application can show the behavior track of the vehicle on the expressway in a chart mode and the like, so that the user can intuitively understand the behavior track and possible problems, and the user is supported to record the analysis result of the behavior track. For example, displaying the possible events of the highway attention area in a special highlight form; the area of interest includes toll booths and gantries, and the event of interest includes known anomalous behavior.

In an exemplary embodiment, the system further comprises a machine learning module, which is used for auditing analysis according to the displayed chart and feeding back auditing information; and after the audit information is fed back, supervised training learning and optimization are carried out according to the abnormal driving behaviors and the simulated driving tracks. In the embodiment of the application, the input becomes a training example based on the information result explicitly fed back by the user. The AI, after obtaining targeted supervised learning, generates a supervised learning algorithm that analyzes the training data and makes its inference to produce an optimal solution, i.e., to allow the algorithm to correctly determine the class label without the label being visible. By receiving result data formed after the abnormal behaviors are analyzed by a user, training and learning are carried out on the result data by using the technologies such as AI machine learning and the like, an AI algorithm model is formed and is put into the abnormal behavior analysis and AI identification abnormality again for application, so that the abnormal behaviors with higher accuracy and (reformed) abnormal types and abnormal probabilities are produced. Based on the feedback information formed after manual checking, new available data analysis logic can be further deduced, and the system or the model can be optimized through modifying program implementation. The training learning system in the application can change the process into an artificial intelligence learning process automatically realized by a system or a model.

In conclusion, the highway has the nationwide networking charging parameters which are universal in the customized industry and the charging modules of each province carry out path reduction and charging on the vehicle passing condition, and the system simplifies the processing mode through data analysis, improves the data processing speed and can meet the requirement of auditing business. The system analyzes mass data by using a big data analysis technology, restores the driving condition of a vehicle on the highway by data and excavates abnormal driving behaviors in the vehicle; meanwhile, a large number of vehicle driving images shot in the target area are analyzed into vehicle image structured data through an AI vision technology, and the memory is saved by storing the vehicle image structured data compared with directly storing the images, so that the storage pressure is reduced. The system can quickly and effectively apply valuable vehicle information in the image to form vehicle identity information ReID independent of license plate recognition; after AI machine learning is introduced, new abnormal behaviors can be found in an unsupervised training mode, and a good closed loop of human-computer cooperation and mutual growth check escape business work can be formed in a supervised training mode. The system can quickly and effectively process million-level data, and suspect behavior data with high accuracy and service value is screened out through data analysis; the data percentage may be 0.37% (the data still belongs to the suspect data, and the fee evasion fact can be confirmed by manual confirmation, and the data can obtain the empirical value for confirming the fee evasion percentage after the customer applies for a certain time). And the system can fuse the highway data and the artificial feedback information to form a self-learning growing closed-loop service flow. Meanwhile, highway charging data are converted into track information records with strong readability and reflecting vehicle behavior characteristics, track data meeting the requirement of the audit service are output through analysis and screening, a feedback result is formed after manual checking, and the result is used as input to conduct training optimization. In the image analysis stage, the system can manually mark the features of the vehicles in the image (such as license plates, head annual inspection labels and the like), and search a result image similar to the features of the target vehicle through feature matching.

The embodiment of the present application further provides a vehicle behavior management device, including:

acquiring traffic transaction data formed when one or more vehicles pass through a target area and vehicle image structured data associated with the traffic transaction data; the vehicle image structured data is obtained by analyzing an image when the vehicle passes through the target area;

and performing behavior management on one or more vehicles based on the traffic transaction data and the vehicle image structured data.

In this embodiment, the vehicle behavior management device executes the method or the system, and specific functions and technical effects may refer to the embodiments, which are not described herein again.

An embodiment of the present application further provides an apparatus, which may include: one or more processors; and one or more machine readable media having instructions stored thereon that, when executed by the one or more processors, cause the apparatus to perform the method of fig. 1. In practical applications, the device may be used as a terminal device, and may also be used as a server, where examples of the terminal device may include: the mobile terminal includes a smart phone, a tablet computer, an electronic book reader, an MP3 (Moving Picture Experts Group Audio Layer III) player, an MP4 (Moving Picture Experts Group Audio Layer IV) player, a laptop, a vehicle-mounted computer, a desktop computer, a set-top box, an intelligent television, a wearable device, and the like.

The present embodiment also provides a non-volatile readable storage medium, where one or more modules (programs) are stored in the storage medium, and when the one or more modules are applied to a device, the device may execute instructions (instructions) included in the data processing method in fig. 1 according to the present embodiment.

Fig. 6 is a schematic diagram of a hardware structure of a terminal device according to an embodiment of the present application. As shown, the terminal device may include: an input device 1100, a first processor 1101, an output device 1102, a first memory 1103, and at least one communication bus 1104. The communication bus 1104 is used to implement communication connections between the elements. The first memory 1103 may include a high-speed RAM memory, and may also include a non-volatile storage NVM, such as at least one disk memory, and the first memory 1103 may store various programs for performing various processing functions and implementing the method steps of the present embodiment.

Alternatively, the first processor 1101 may be, for example, a Central Processing Unit (CPU), an Application Specific Integrated Circuit (ASIC), a Digital Signal Processor (DSP), a Digital Signal Processing Device (DSPD), a Programmable Logic Device (PLD), a Field Programmable Gate Array (FPGA), a controller, a microcontroller, a microprocessor, or other electronic components, and the processor 1101 is coupled to the input device 1100 and the output device 1102 through a wired or wireless connection.

Optionally, the input device 1100 may include a variety of input devices, such as at least one of a user-oriented user interface, a device-oriented device interface, a software programmable interface, a camera, and a sensor. Optionally, the device interface facing the device may be a wired interface for data transmission between devices, or may be a hardware plug-in interface (e.g., a USB interface, a serial port, etc.) for data transmission between devices; optionally, the user-facing user interface may be, for example, a user-facing control key, a voice input device for receiving voice input, and a touch sensing device (e.g., a touch screen with a touch sensing function, a touch pad, etc.) for receiving user touch input; optionally, the programmable interface of the software may be, for example, an entry for a user to edit or modify a program, such as an input pin interface or an input interface of a chip; the output devices 1102 may include output devices such as a display, audio, and the like.

In this embodiment, the processor of the terminal device includes a function for executing each module of the speech recognition apparatus in each device, and specific functions and technical effects may refer to the above embodiments, which are not described herein again.

Fig. 7 is a schematic hardware structure diagram of a terminal device according to another embodiment of the present application. FIG. 7 is a specific embodiment of the implementation of FIG. 6. As shown, the terminal device of the present embodiment may include a second processor 1201 and a second memory 1202.

The second processor 1201 executes the computer program code stored in the second memory 1202 to implement the method described in fig. 1 in the above embodiment.

The second memory 1202 is configured to store various types of data to support operations at the terminal device. Examples of such data include instructions for any application or method operating on the terminal device, such as messages, pictures, videos, and so forth. The second memory 1202 may include a Random Access Memory (RAM) and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

Optionally, a second processor 1201 is provided in the processing assembly 1200. The terminal device may further include: communication components 1203, power components 1204, multimedia components 1205, audio components 1206, input/output interfaces 1207, and/or sensor components 1208. The specific components included in the terminal device are set according to actual requirements, which is not limited in this embodiment.

The processing component 1200 generally controls the overall operation of the terminal device. The processing assembly 1200 may include one or more second processors 1201 to execute instructions to perform all or part of the steps of the method illustrated in fig. 1 described above. Further, the processing component 1200 can include one or more modules that facilitate interaction between the processing component 1200 and other components. For example, the processing component 1200 can include a multimedia module to facilitate interaction between the multimedia component 1205 and the processing component 1200.

The power supply component 1204 provides power to the various components of the terminal device. The power components 1204 may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for the terminal device.

The multimedia components 1205 include a display screen that provides an output interface between the terminal device and the user. In some embodiments, the display screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the display screen includes a touch panel, the display screen may be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation.

The audio component 1206 is configured to output and/or input speech signals. For example, the audio component 1206 includes a Microphone (MIC) configured to receive external voice signals when the terminal device is in an operational mode, such as a voice recognition mode. The received speech signal may further be stored in the second memory 1202 or transmitted via the communication component 1203. In some embodiments, audio component 1206 also includes a speaker for outputting voice signals.

The input/output interface 1207 provides an interface between the processing component 1200 and peripheral interface modules, which may be click wheels, buttons, etc. These buttons may include, but are not limited to: a volume button, a start button, and a lock button.

The sensor component 1208 includes one or more sensors for providing various aspects of status assessment for the terminal device. For example, the sensor component 1208 may detect an open/closed state of the terminal device, relative positioning of the components, presence or absence of user contact with the terminal device. The sensor assembly 1208 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact, including detecting the distance between the user and the terminal device. In some embodiments, the sensor assembly 1208 may also include a camera or the like.

The communication component 1203 is configured to facilitate communications between the terminal device and other devices in a wired or wireless manner. The terminal device may access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In one embodiment, the terminal device may include a SIM card slot therein for inserting a SIM card therein, so that the terminal device may log onto a GPRS network to establish communication with the server via the internet.

As can be seen from the above, the communication component 1203, the audio component 1206, the input/output interface 1207 and the sensor component 1208 in the embodiment of fig. 7 may be implemented as the input device in the embodiment of fig. 6.

The foregoing embodiments are merely illustrative of the principles and utilities of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.

Claims

1. A vehicle behavior management method characterized by comprising the steps of:

2. The vehicle behavior management method according to claim 1, characterized in that the behavior management of the one or more vehicles includes at least: analyzing whether abnormal driving behaviors exist in the one or more vehicles.

3. The vehicle behavior management method according to claim 2, characterized by further comprising:

4. The vehicle behavior management method according to claim 3, wherein the specific process of analyzing the recovered driving trajectory according to the pre-constructed highway network model to determine whether the one or more vehicles have abnormal driving behaviors comprises:

5. The vehicle behavior management method according to any one of claims 2 to 4, characterized in that, if there is an abnormal traveling behavior of a certain vehicle, the traveling behavior of the vehicle includes at least one of: the method comprises the steps that corresponding passing transaction data are not generated when a vehicle passes through a target area, mileage transaction amount when the vehicle generates the passing transaction data is not matched with corresponding driving mileage, the type of a toll vehicle is not consistent with a preset billing vehicle type when the vehicle generates the passing transaction data, the vehicle drives in two opposite directions in a single driving track, a plurality of vehicle-mounted electronic label device sets exist in the single driving track, and a plurality of composite pass cards exist in the single driving track.

6. The vehicle behavior management method according to claim 1, wherein the specific process of acquiring the vehicle image structured data includes:

7. The vehicle behavior management method according to claim 3, further comprising, before restoring the trajectory of the one or more vehicles, cleaning the traffic transaction data according to highway common sense and business rules; and separating, combining and cleaning the traffic transaction data and the license plate information according to the highway network rule.

8. The vehicle behavior management method according to claim 4, further comprising inputting a target image containing a target vehicle; searching out similar images meeting a preset similarity range from images of a vehicle passing through a target area through a visual intelligent engine, and displaying the similar images in an image list view; wherein the displayed information at least comprises image time and image similarity.

9. The vehicle behavior management method according to claim 4 or 8, characterized by further comprising inputting a target image containing a target vehicle; searching out similar images meeting a preset similarity range from images of a vehicle passing through a target area through a visual intelligent engine; and determining the running track of the target vehicle within the inquiry time range according to the similar image, and displaying the running track.

10. The vehicle behavior management method according to claim 6, wherein the license plate information includes at least: license plate number and license plate color;

the vehicle basic attributes at least include: vehicle brand and vehicle type;

the vehicle service attributes include at least: and (4) hazardous chemicals.

11. The vehicle behavior management method according to claim 4, further comprising displaying the vehicle abnormal running behavior analysis result, the restored running track, and the simulated running track in a visualized graph;

12. The vehicle behavior management method according to claim 11, further comprising performing audit analysis based on the displayed chart, feeding back audit information; and after the audit information is fed back, supervised training learning and optimization are carried out according to the abnormal driving behaviors and the simulated driving tracks of the vehicle.

13. The vehicle behavior management method according to claim 1, characterized in that the target area includes at least one of: toll stations at the entrance of the highway, toll stations at the exit of the highway and a portal frame between the entrance and the exit of the highway.

14. A vehicle behavior management system, comprising:

15. The vehicle behavior management system of claim 14, wherein the behavior management of the one or more vehicles by the vehicle behavior management module includes analyzing whether abnormal driving behavior exists for the one or more vehicles; then there are:

16. The vehicle behavior management system according to claim 15, wherein the vehicle behavior management module analyzes the recovered driving trajectory according to a pre-constructed highway network model, and the specific process of determining whether the one or more vehicles have abnormal driving behaviors comprises:

17. The vehicle behavior management system according to claim 15 or 16, characterized in that, if there is an abnormal traveling behavior of a certain vehicle, the traveling behavior of the certain vehicle includes at least one of: the method comprises the steps that corresponding passing transaction data are not generated when a vehicle passes through a target area, mileage transaction amount when the vehicle generates the passing transaction data is not matched with corresponding driving mileage, the type of a toll vehicle is not consistent with a preset billing vehicle type when the vehicle generates the passing transaction data, the vehicle drives in two opposite directions in a single driving track, a plurality of vehicle-mounted electronic label device sets exist in the single driving track, and a plurality of composite pass cards exist in the single driving track.

18. The vehicle behavior management system according to claim 15, further comprising a first search module for inputting a target image containing a target vehicle; searching out similar images meeting a preset similarity range from images of a vehicle passing through a target area through a visual intelligent engine, and displaying the similar images in an image list view; wherein the displayed information at least comprises image time and image similarity.

19. The vehicle behavior management system according to claim 15 or 18, further comprising a second search module for inputting a target image containing a target vehicle; searching out similar images meeting a preset similarity range from images of a vehicle passing through a target area through a visual intelligent engine; and determining the running track of the target vehicle within the inquiry time range according to the similar image, and displaying the running track.

20. A vehicle behavior management apparatus characterized by comprising:

21. An apparatus, comprising:

one or more processors; and

one or more machine-readable media having instructions stored thereon that, when executed by the one or more processors, cause the apparatus to perform the method of any of claims 1-13.

22. One or more machine readable media having instructions stored thereon that, when executed by one or more processors, cause an apparatus to perform the method of any of claims 1-13.