WO2017125063A1

WO2017125063A1 - Processing method and device for vehicle traffic violation

Info

Publication number: WO2017125063A1
Application number: PCT/CN2017/071833
Authority: WO
Inventors: 王勃飞; 郑成建; 刘宇
Original assignee: 中兴通讯股份有限公司
Priority date: 2016-01-20
Filing date: 2017-01-20
Publication date: 2017-07-27
Also published as: CN106991820A; CN106991820B

Abstract

A processing method and device for vehicle traffic violation. The method comprises: detecting the position and type of a vehicle in a surveillance video within a predetermined period (S102); determining, according to the detected position and type and a predetermined relationship between vehicle types and legal lane usage, whether the vehicle is in a lane corresponding to the detected type (S104); and if not, then determining the vehicle as a vehicle in violation of rules (S106). The above technical solution solves the problem in the related art in which automated detection of illegal lane usage of a vehicle cannot be achieved in a scenario in which different types of vehicles correspond to different lanes, thus realizing automated detection of illegal lane usage for different types of vehicles.

Description

Violating vehicle processing method and device

Technical field

This article relates to, but is not limited to, the field of intelligent transportation technology, and relates to a method and device for handling illegal vehicles.

Background technique

Vehicle illegal occupation is a typical dynamic traffic violation. This illegal behavior not only seriously affects the normal traffic order of the expressway, but also reduces the traffic efficiency, and the road traffic accidents caused by it are increasing year by year, which leads to malignancy. The main cause of frequent road traffic accidents. The dynamic illegal behavior during such driving is difficult to manage by manpower, and has become a difficult task in highway traffic management in China.

In recent years, with the development of computer image processing and recognition technology, the use of machine vision instead of artificial vision for target extraction and recognition, and the automatic detection of illegal behaviors that do not follow the prescribed lanes have emerged. The traditional method of manpower has greatly improved the automatic detection capability of illegal traffic occupation of vehicles, and it is an effective means to combat the illegal occupation of vehicles.

In the related art, a method for detecting illegal lanes of a vehicle is provided, which includes: extracting an image including a lane in a video frame; determining whether a lane in the image is a preset color lane; and when the lane is a preset color lane, in the video Detecting a license plate in the frame; when detecting the license plate, performing color discrimination on the license plate to determine whether the color of the license plate is a limited color; and when the color of the license plate is a limited color, detecting whether the license plate is in the predetermined The color lane is set; when the license plate is in the preset color lane, the license plate is marked in the video frame. In order to enable the relevant departments to dispose of social vehicles occupying bus lanes or other dedicated lanes according to the marked license plates, thereby improving the operational efficiency and safety of public buses or other vehicles.

However, it should be noted that the above-mentioned method for detecting the illegal occupation of vehicles has certain limitations. The method can only detect the license plates and lanes with special colors, and for different types of vehicles corresponding to different driving lanes. Automatic detection is not possible.

It cannot be realized in the related art in the scenario that different types of vehicles correspond to different driving lanes. The problem of automatic detection of illegal traffic occupation of vehicles has not yet proposed an effective solution.

Summary of the invention

The following is an overview of the topics detailed in this document. This Summary is not intended to limit the scope of the claims.

The embodiment of the invention provides a method and a device for processing an illegal vehicle, which solves the problem that the automatic detection of the illegal occupation of the vehicle cannot be realized in the related art in the scenario that different types of vehicles correspond to different driving lanes.

An embodiment of the present invention provides a method for processing a violating vehicle, comprising: detecting a position and a type of a vehicle in a video during a predetermined time; and determining a predetermined relationship between the vehicle type and the legal driving lane according to the detected location and the type And determining whether the vehicle is located in the detected driving lane corresponding to the type; and if the determination result is no, determining that the vehicle is a violating vehicle.

Optionally, before detecting the location and type of the vehicle in the video for a predetermined time, the method further includes: performing foreground extraction on the monitoring video, obtaining a foreground vehicle image, and recording the vehicle in the foreground vehicle image. Positioning: identifying the foreground vehicle picture, obtaining a vehicle type corresponding to the vehicle in the foreground vehicle picture; performing statistics on the position and type of the vehicle within a predetermined time, and determining a legal driving lane corresponding to each type of vehicle.

Optionally, counting the location and type of the vehicle in the predetermined time, determining the legal driving lane corresponding to each type of vehicle includes: recording the location of the plurality of vehicles in the predetermined time; The positions are grouped according to the type of the vehicle, wherein the positions of the same type of vehicles are placed in the same group; the lanes in which the concentrated areas of the vehicles in each group are located are determined as the legal driving lanes of each type of vehicle.

Optionally, before determining the lane in which the concentrated area of the location of the vehicle in each group is located as the legal driving lane of each type of vehicle, the method further includes: excluding the location of the vehicle in each group is located in the concentrated area Vehicles outside the driveway.

Optionally, after determining that the vehicle is the offending vehicle, the method further includes: identifying vehicle information of the offending vehicle, wherein the vehicle information includes a license plate number and a vehicle body color; and recording the illegal vehicle Vehicle summary consisting of vehicle information, vehicle type, and vehicle snapshot pictures information.

An embodiment of the present invention further provides an illegal vehicle processing apparatus, comprising: a detecting module configured to detect a position and a type of a vehicle in a monitoring video within a predetermined time; and a determining module configured to detect the position and the type according to the detected And determining a predetermined relationship between the vehicle type and the legal driving lane, determining whether the vehicle is located in the detected driving lane corresponding to the type; and determining, configured to determine that the vehicle is in violation if the determination result is negative vehicle.

Optionally, the device further includes: a first recording module, configured to perform foreground extraction on the monitoring video, obtain a foreground vehicle picture, and record a location of the vehicle in the foreground vehicle picture; the first identification module, setting In order to identify the foreground vehicle picture, obtain the vehicle type corresponding to the vehicle in the foreground vehicle picture; the statistical determination module is configured to perform statistics on the position and type of the vehicle within a predetermined time, and determine the legality corresponding to each type of vehicle. Driving lane.

Optionally, the statistical determination module includes: a recording unit configured to record a location of the plurality of vehicles within the predetermined time;

a grouping unit configured to group the locations of the plurality of vehicles according to the vehicle type, wherein the locations of the same type of vehicles are placed in the same group; and the determining unit is configured to determine the lane in which the concentrated area of the vehicle in each group is located The legal driving lane for each type of vehicle.

Optionally, the apparatus further comprises: an exclusion unit configured to exclude vehicles in each group from where the location of the vehicle is located outside the lane in which the concentrated area is located.

Optionally, the device further includes: a second identification module configured to identify vehicle information of the offending vehicle, wherein the vehicle information includes a license plate number and a vehicle body color; and the second recording module is configured to record the violation Vehicle summary information consisting of vehicle information, vehicle type, and vehicle snapshot pictures.

The embodiment of the invention further provides a computer readable storage medium, wherein the computer readable storage medium stores computer executable instructions, and when the computer executable instructions are executed, the offending vehicle processing method is implemented.

The embodiment of the present invention detects the position and type of the vehicle in the video by detecting the predetermined time; and determines whether the vehicle is located according to the detected position and the type, and the predetermined relationship between the vehicle type and the legal driving lane. The type corresponds to the driving lane; the judgment result is Otherwise, the vehicle is determined to be a violating vehicle. The above technical solution solves the problem that the automatic detection of the illegal occupation of the vehicle cannot be realized in the scenario that different types of vehicles correspond to different driving lanes in the related art, and the automatic detection of illegal occupation of different types of vehicles is realized. Other aspects will be apparent upon reading and understanding the drawings and detailed description.

DRAWINGS

The drawings described herein are intended to provide a further understanding of the present application, and are intended to be a part of this application. In the drawing:

1 is a flowchart of a method for processing a violating vehicle according to Embodiment 1 of the present invention;

2 is a first schematic structural view of a vehicle handling device for violation of a vehicle according to Embodiment 1 of the present invention;

3 is a second block diagram of a vehicle handling device for violation of the first embodiment of the present invention;

Figure 4 is a block diagram 3 of the illegal vehicle processing apparatus according to the first embodiment of the present invention;

Figure 5 is a block diagram 4 of the illegal vehicle processing apparatus according to the first embodiment of the present invention;

6 is a flow chart of a method for detecting a vehicle that does not follow a prescribed lane according to a second embodiment of the present invention;

7 is a flow chart of automatically detecting a legal driving lane corresponding to each type of vehicle in a traffic monitoring video according to Embodiment 2 of the present invention;

8 is a flow chart of detecting a vehicle that does not travel according to regulations according to Embodiment 2 of the present invention;

FIG. 9 is a structural block diagram of an adaptive vehicle driving device that does not follow a prescribed lane according to a second embodiment of the present invention.

detailed description

The present application will be described in detail below with reference to the drawings in conjunction with the embodiments. It should be noted that the embodiments in the present application and the features in the embodiments may be combined with each other without conflict.

Embodiment 1

An embodiment of the present invention provides a method for processing an illegal vehicle, and FIG. 1 is an embodiment according to the present invention. A flow chart of a method for handling a violating vehicle, as shown in Figure 1, includes:

Step S102, detecting a position and a type of the vehicle in the monitoring video within a predetermined time;

Step S104, determining, according to the detected position and the type, and the predetermined relationship between the vehicle type and the legal driving lane, whether the vehicle is located in the detected driving lane corresponding to the type;

In step S106, if the determination result is no, it is determined that the vehicle is a violating vehicle.

Through the above technical solution, detecting the position and type of the vehicle in the monitoring video within a predetermined time; determining whether the vehicle is located in the detected type according to the detected position and the type, and the predetermined relationship between the vehicle type and the legal driving lane In the case of the driving lane, if the determination result is no, the vehicle is determined to be a violating vehicle, and the problem that the automatic detection of the illegal lane occupation of the vehicle cannot be realized in the related art for different types of vehicles corresponding to different driving lanes is realized. Automatically detect illegal occupation of different types of vehicles.

In this embodiment, when the monitoring of the video is started, the monitoring video may be learned to determine the legal driving lane of each type of vehicle. Before monitoring the location and type of the vehicle in the video for a predetermined time, the method may further include: Performing foreground extraction on the surveillance video, obtaining a foreground vehicle picture, and recording a location of the vehicle in the foreground vehicle picture; identifying the foreground vehicle picture to obtain a vehicle type corresponding to the vehicle in the foreground vehicle picture; The location and type are counted to determine the legal driving lane for each type of vehicle.

Optionally, the location and type of the vehicle are counted for the predetermined time, and determining the legal driving lane corresponding to each type of the vehicle may include:

Recording the location of the plurality of vehicles within the predetermined time; grouping the locations of the plurality of vehicles according to the vehicle type, wherein the locations of the same type of vehicles are placed in the same group; the concentrated areas of the locations where the vehicles are located in each group are located The lane is determined as the legal lane of travel for each type of vehicle.

Since in the process of learning the legal driving lanes corresponding to different types of vehicles, there is a possibility that illegally occupying vehicles exist, and vehicles that are illegally driven at this time should be eliminated. Therefore, the concentration of the vehicles in each group is concentrated. Before the lane in which the zone is located is determined as the legal driving lane of each type of vehicle, the vehicles in each group whose vehicles are located outside the lane in which the concentrated area is located may be excluded.

In order to facilitate the recording, after determining that the vehicle is the illegal vehicle, the method may further include: The vehicle information of the vehicle is identified, wherein the vehicle information includes a license plate number and a vehicle body color; vehicle summary information composed of the vehicle information of the illegal vehicle, the vehicle type, and the vehicle captured picture is recorded.

The embodiment of the present invention further provides a device for violating the vehicle, and FIG. 2 is a schematic structural diagram 1 of the device for violating the vehicle according to the embodiment of the present invention. As shown in FIG. 2, the method includes:

The detecting module 22 is configured to detect a position and a type of the vehicle in the monitoring video within a predetermined time;

The determining module 24 is configured to determine, according to the detected position and the type, and the predetermined relationship between the vehicle type and the legal driving lane, whether the vehicle is located in the detected driving lane corresponding to the type;

The determination module 26 is arranged to determine that the vehicle is a violating vehicle if the determination result is negative.

Optionally, FIG. 3 is a schematic structural diagram 2 of the illegal vehicle processing apparatus according to the embodiment of the present invention. As shown in FIG. 3, the apparatus further includes:

The first recording module 32 is configured to perform foreground extraction on the surveillance video, obtain a foreground vehicle image, and record a location of the vehicle in the foreground vehicle image;

The first identification module 34 is configured to identify the foreground vehicle image, and obtain a vehicle type corresponding to the vehicle in the foreground vehicle image;

The statistical determination module 36 is configured to count the position and type of the vehicle within a predetermined time to determine a legal driving lane corresponding to each type of vehicle.

Optionally, FIG. 4 is a schematic structural diagram 3 of the illegal vehicle processing apparatus according to the embodiment of the present invention. As shown in FIG. 4, the statistical determination module 36 includes:

a recording unit 42 configured to record a location of the plurality of vehicles within the predetermined time;

The grouping unit 44 is configured to group the locations of the plurality of vehicles according to the vehicle type, wherein the locations of the same type of vehicles are placed in the same group;

The determining unit 46 is arranged to determine the lane in which the concentrated area of the vehicle in each group is located as the legal driving lane of each type of vehicle.

Optionally, FIG. 5 is a schematic structural diagram 4 of the illegal vehicle processing apparatus according to the embodiment of the present invention, such as As shown in FIG. 5, the device further includes:

a second identification module 52, configured to identify vehicle information of the offending vehicle, wherein the vehicle information includes a license plate number and a vehicle body color;

The second recording module 54 is configured to record vehicle information of the illegal vehicle, a vehicle type, and vehicle summary information composed of a vehicle captured picture.

Embodiment 2

The embodiments of the present invention are further described below in conjunction with specific embodiments.

The embodiment of the invention provides an adaptive vehicle detection method that does not follow the prescribed lane. By analyzing the traffic monitoring video, the legal driving lane corresponding to each type of vehicle in the traffic monitoring video is automatically detected, and then the detected lane is determined according to the detected lane. The specified vehicle type corresponding to the area and the lane area identifies the vehicle type in the lane area, detects the vehicle that does not travel according to the regulations, and finally further identifies the illegally driven vehicle, and records the vehicle summary information for easy investigation. Forensics. Compared with related schemes and technologies, automatic processing and analysis can be realized without manual intervention in specifying the lane area of each type of vehicle, and it has the characteristics of high precision and high real-time.

FIG. 6 is a flowchart of a method for detecting a vehicle that does not follow a prescribed lane according to an embodiment of the present invention. As shown in FIG. 6, the method may include:

Step S602, analyzing the traffic monitoring video, and automatically detecting the legal driving lane corresponding to each type of vehicle in the traffic monitoring video;

Step S604, identifying the vehicle type in the lane region based on the detected vehicle type and the predetermined vehicle type corresponding to the lane region, and detecting the vehicle that does not travel according to the regulations.

FIG. 7 is a flowchart of automatically detecting a legal driving lane corresponding to each type of vehicle in a traffic monitoring video according to an embodiment of the present invention, as shown in FIG. 7, including:

Step S702, using a Hough transform based method to perform lane line detection on the image to be processed, and obtaining all lane line coordinates in the image;

Step S704, using a Gaussian Mixture Model (GMM) to perform foreground extraction on the image to be processed, obtaining a foreground vehicle image, and recording the location of the foreground vehicle;

Step S706, using a sparsely encoded scale-invariant feature transform (SIFT) feature combined with a support vector machine (SVM) classifier to identify the foreground vehicle image. Get the vehicle type corresponding to the foreground vehicle image:

Step S708, performing statistics on the foreground vehicle position and type within a predetermined time (eg, 10 minutes) to obtain a legal driving lane corresponding to each type of vehicle:

Assuming that a total of 200 vehicles are detected within 10 minutes, the positions of the 200 vehicles are recorded as pos_1, pos_2, ..., pos_200 according to the output of the foreground detection method, where for each position pos_i(i∈[1,200]) The vehicle center position at the position is described by the (abscissa, ordinate) point pair (pos_i_x, pos_i_y) in the surveillance video image. According to the vehicle type identification method, the types of the 200 vehicles are recorded as class_1, class_2, ... ..., class_200, where class_i(i∈[1,200]) represents the vehicle type identified by the vehicle identification method (assuming there are three types of vehicles, such as cars, transportation vehicles, trucks); The vehicle types are grouped and the same type of vehicle is placed in the same group. Assuming that the grouping, the result of the jth group is expressed as the group_j(j∈[1,3])={pos_i|i∈[1,200] and class_i==j}.

Assuming that the car is traveling in the direction of the ordinate axis, each group_j (j∈[1,3]) is processed as follows:

Exclude the outliers in the direction of the abscissa axis in group_j, that is, the points farther away from the mean points of all the positions in the direction of the abscissa axis (these points may contain illegally driven vehicles); fit out the remaining positions in group_j A straight line segment is recorded as lineSeg_j. The inner area of the lane line detected by the left and right sides of lineSeg_j is the legal driving lane of the type of vehicle, and is recorded as driveway_j.

FIG. 8 is a flowchart of detecting a vehicle that does not travel according to regulations according to an embodiment of the present invention. As shown in FIG. 8, the method includes:

Step S802, using a Gaussian Mixture Model (GMM) to perform foreground extraction on the image to be processed, obtaining a foreground vehicle image, and recording an image area covered by the vehicle;

Step S804, using a sparsely encoded scale-invariant feature transform (SIFT) feature combined with a support vector machine (SVM) classifier to identify the foreground vehicle image. Obtaining the vehicle type corresponding to the foreground vehicle picture;

Step S806, calculating whether the image area covered by the vehicle completely belongs to a legal lane corresponding to the vehicle of the type, and if the image area covered by the vehicle is not a legal lane corresponding to the vehicle of the type, the vehicle is considered to be driving illegally;

Assuming that the currently processed vehicle number is 100 and the vehicle type identification result class_100=2, the legal lane of the vehicle of this type is driveway_2, wherein driveway_2 is described by the binary template image binmask_of_driveway_2 of the area covered by the left and right lane lines of the lane area, ie The pixel coordinates (x, y) belonging to the region satisfy binmask_of_driveway_2(x, y) = 1, and conversely, the pixel coordinates (x, y) binmask_of_driveway_2(x, y) = 0 that do not belong to the region. Assume that the coverage area of the currently processed vehicle is described by the upper left vertex (x0, y0), the upper right vertex (x1, y1), the lower left vertex (x2, y2), and the lower right vertex (x3, y3) of the circumscribed rectangle of the area; The coordinates are all within the range of the driveway_2 lane, that is, any i∈[0,3] satisfies binmask_of_driveway_2(xi,yi)=1, then the image area covered by the vehicle is considered to belong to the legal lane corresponding to the type of vehicle; otherwise, If the above four coordinates are not all within the range of the driveway_2 lane, that is, the existing i∈[0,3] satisfies binmask_of_driveway_2(xi,yi)=0, the vehicle is considered to be driving illegally.

In an optional embodiment, the illegally traveling vehicle can be further identified, including license plate recognition, vehicle body color recognition, and recording the vehicle summary information composed of the vehicle type and the vehicle captured image. It is easy to investigate and collect evidence.

FIG. 9 is a schematic structural diagram of an adaptive vehicle driving device that does not follow a prescribed lane according to an embodiment of the present invention. As shown in FIG. 9, the device may include: an analyzing unit 92, a detecting unit 94, and a recording unit 96, wherein The function of the unit 92 is implemented by the first recording module 32, the first identification module 34 and the statistical determination module 36. The function of the detecting unit 94 is implemented by the detecting module 22, the determining module 24 and the determining module 26, and the recording unit 96 features by the above The two identification modules 62 are implemented together with the second recording module 64, each of which is further described below.

The analyzing unit 92 is configured to analyze the traffic monitoring video, and automatically detect the legal driving lane corresponding to each type of vehicle in the traffic monitoring video;

The detecting unit 94 is configured to identify the vehicle type in the lane area according to the detected vehicle area and the specified vehicle type corresponding to the lane area, and detect the vehicle that does not travel according to the regulations;

The recording unit 96 is configured to further identify the illegally traveling vehicle, including license plate recognition, vehicle body color recognition, and record the vehicle identification information composed of the vehicle type and the vehicle captured image, so as to facilitate investigation and forensics. .

Optionally, the analyzing unit 92 may include a lane detecting subunit 922, a foreground detecting subunit 924, a vehicle type identifying subunit 926, and a statistical analysis subunit 928, where:

The lane detection sub-unit 922 is configured to detect lane lines in the image to be processed, and obtain all lane line position coordinates;

The foreground detection sub-unit 924 is configured to perform foreground extraction on the image to be processed, obtain a foreground vehicle picture, and record the location of the foreground vehicle;

The vehicle identification unit 926 is configured to identify the foreground vehicle picture to obtain a vehicle type corresponding to the foreground vehicle picture;

The statistical analysis sub-unit 928 is configured to count the foreground vehicle position and type within a certain time range to obtain a legal driving lane corresponding to each type of vehicle.

Optionally, the detecting unit 94 may include a vehicle area acquiring subunit 942, a vehicle type identifying subunit 944, and a discriminating subunit 946, where:

The vehicle area acquisition subunit 942 is configured to perform foreground extraction on the image to be processed, obtain a foreground vehicle picture, and record an image area covered by the vehicle;

The vehicle identification unit 944 is configured to identify the foreground vehicle picture to obtain a vehicle type corresponding to the foreground vehicle picture;

The discriminating sub-unit 946 is configured to calculate whether the image area covered by the vehicle completely belongs to a legal lane corresponding to the vehicle of the type. If the image area covered by the vehicle does not belong to a legal lane corresponding to the vehicle of the type, the vehicle is considered to be driving illegally.

In the embodiment of the present invention, the traffic monitoring video is analyzed by the analyzing unit 92, and the legal driving lane corresponding to each type of vehicle in the traffic monitoring video is automatically detected, and then detected by the detecting unit 94 according to the detected lane region and the lane region. The vehicle type is specified to identify the vehicle type in the lane area, and the vehicle that does not travel according to the regulations is detected. Finally, the vehicle that violates the driving is further identified by the recording unit 96, and the vehicle summary information is recorded, so as to facilitate investigation and evidence collection. The above technical solution can realize full-automatic processing and analysis, and has the characteristics of high precision and high real-time performance.

The above description is only an optional embodiment of the present application, and is not intended to limit the present application, and various changes and modifications may be made to the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of this application are intended to be included within the scope of the present application.

One of ordinary skill in the art will appreciate that all or a portion of the above steps may be performed by a program to instruct related hardware, such as a processor, which may be stored in a computer readable storage medium, such as a read only memory, disk or optical disk. Wait. Alternatively, all or part of the steps of the above embodiments may also be implemented using one or more integrated circuits. Correspondingly, each module/unit in the above embodiment may be implemented in the form of hardware, for example, by implementing an integrated circuit to implement its corresponding function, or may be implemented in the form of a software function module, for example, executing a program stored in the memory by a processor. / instruction to achieve its corresponding function. This application is not limited to any specific combination of hardware and software. A person skilled in the art should understand that the technical solutions of the present application can be modified or equivalent, without departing from the spirit and scope of the technical solutions of the present application, and should be included in the scope of the claims of the present application.

Industrial applicability

The above technical solution solves the problem that the automatic detection of the illegal occupation of the vehicle cannot be realized in the scenario that different types of vehicles correspond to different driving lanes in the related art, and the automatic detection of illegal occupation of different types of vehicles is realized.

Claims

A method for handling illegal vehicles, including:

Detecting the location and type of the vehicle in the surveillance video within a predetermined time period;

Determining, according to the detected position and the type, and a predetermined relationship between the vehicle type and the legal driving lane, whether the vehicle is located in the detected driving lane corresponding to the type;

In the case where the determination result is negative, it is determined that the vehicle is a violating vehicle.
The method of claim 1, before detecting the location and type of the vehicle in the video for a predetermined time, the method further comprising:

Performing foreground extraction on the monitoring video, obtaining a foreground vehicle picture, and recording a location of the vehicle in the foreground vehicle picture;

Identifying the foreground vehicle picture to obtain a vehicle type corresponding to the vehicle in the foreground vehicle picture;

The location and type of the vehicle are counted for a predetermined time to determine the legal driving lane corresponding to each type of vehicle.
The method according to claim 2, wherein: counting the position and type of the vehicle within the predetermined time, determining the legal driving lane corresponding to each type of vehicle comprises:

Recording the location of the plurality of vehicles within the predetermined time;

Grouping the locations of multiple vehicles according to vehicle type, wherein the locations of the same type of vehicles are placed in the same group;

The lane in which the concentrated area of the vehicle in each group is located is determined as the legal driving lane of each type of vehicle.
The method according to claim 3, before determining the lane in which the concentrated area of the vehicle is located in each group is determined as the legal driving lane of each type of vehicle, the method further comprising:

Vehicles in each group whose vehicles are located outside the lane in which the concentrated area is located are excluded.
The method of any one of claims 1 to 4, after determining that the vehicle is the offending vehicle, the method further comprising:

Identifying vehicle information of the offending vehicle, wherein the vehicle information includes a license plate number and the color of car;

Recording vehicle information of the illegal vehicle, vehicle type, and vehicle summary information composed of the vehicle snapshot picture.
An illegal vehicle handling device comprising:

a detecting module configured to detect a position and a type of the vehicle in the monitoring video within a predetermined time;

a determining module, configured to determine, according to the detected position and the type, and a predetermined relationship between the vehicle type and the legal driving lane, whether the vehicle is located in the detected driving lane corresponding to the type;

The determining module is configured to determine that the vehicle is a violating vehicle if the determination result is negative.
The apparatus of claim 6 further comprising:

a first recording module, configured to perform foreground extraction on the monitoring video, obtain a foreground vehicle image, and record a location of the vehicle in the foreground vehicle image;

a first identification module configured to identify the foreground vehicle picture to obtain a vehicle type corresponding to the vehicle in the foreground vehicle picture;

The statistical determination module is configured to perform statistics on the position and type of the vehicle within a predetermined time to determine a legal driving lane corresponding to each type of vehicle.
The apparatus of claim 7 wherein: said statistical determination module comprises:

a recording unit configured to record a location of the plurality of vehicles within the predetermined time period;

a grouping unit, configured to group the locations of the plurality of vehicles according to the vehicle type, wherein the locations of the same type of vehicles are placed in the same group;

The determining unit is configured to determine the lane in which the concentrated area of the vehicle in each group is located as the legal driving lane of each type of vehicle.
The apparatus of claim 8 further comprising:

The exclusion unit is arranged to exclude vehicles in each group from where the vehicle is located outside the lane in which the concentrated area is located.
The apparatus according to any one of claims 6 to 9, the apparatus further comprising:

a second identification module configured to identify vehicle information of the offending vehicle, wherein the vehicle Vehicle information includes license plate number and body color;

The second recording module is configured to record vehicle information of the illegal vehicle, a vehicle type, and vehicle summary information composed of a vehicle captured picture.