A vehicle violation identification method, device, binocular camera and storage medium

The field of view of the binocular camera covers the signs and traffic intersections of variable lanes, and combines the vehicle characteristics and driving direction characteristics to accurately identify vehicle violations in variable lanes, solving the existing technology of vehicles in variable lanes Issues of accuracy of violation identification.

CN118397577BActive Publication Date: 2024-12-03HANGZHOU HIKVISION DIGITAL TECHNOLOGY CO LTD
View PDF 14 Cites 0 Cited by

Patent Information

Application Number
CN202410840406.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-06-26
Publication Date
2024-12-03
Estimated Expiration
2044-06-26

AI Technical Summary

Technical Problem

The prior art is difficult to accurately identify whether vehicles in variable lanes have violated their behavior, especially when the direction of travel of variable lanes in urban roads changes dynamically with traffic or time.

Method used

Using a binocular camera, image acquisition and analysis are carried out through the first camera and the second camera respectively corresponding to their field of view, and the vehicle's vehicle characteristics, lane number and driving direction characteristics are obtained, and combined with the sign characteristics, determine whether the vehicle exists without guidance. Driving violations.

Benefits of technology

Accurate identification of vehicle violations in variable lanes is achieved, the accuracy and efficiency of identification is improved, and the dispute and waste rate is reduced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118397577B_ABST
    Figure CN118397577B_ABST
Patent Text Reader

Abstract

The embodiment of the present application provides a method, device, binocular camera and storage medium for identifying vehicle violations, which relates to the field of traffic management technology. The method includes: controlling a first camera to acquire images, and analyzing the acquired images to obtain a first feature associated with the vehicle; controlling a second camera to acquire images, and analyzing the acquired images to obtain a second feature associated with the vehicle; determining a vehicle that is the same vehicle as the target vehicle from vehicles associated with the first feature, and using the first feature of the determined vehicle as the first reference feature; in response to the first reference feature being the same as the second feature lane number, using the driving direction indicated by the sign feature as the basis for judgment, and based on the driving direction feature in the second feature associated with the target vehicle, judging whether the target vehicle has violated the regulations by not driving according to the guidance. This solution can accurately identify whether a vehicle in a variable lane has violated regulations.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of traffic management technology, and in particular to a method, device, binocular camera and storage medium for identifying vehicle violations. Background Art

[0002] Vehicles must meet driving regulations during driving, for example, they must follow lane guidance, pass intersections according to traffic lights, park according to regulations, change lanes according to regulations, etc. In order to reduce vehicle violations, it is usually necessary to identify vehicle violations.

[0003] Currently, cameras are usually installed at traffic intersections to identify the behavior of vehicles within the traffic intersection and determine whether the vehicle has committed any violations such as not following the lane guidance.

[0004] Current cameras usually detect vehicles in fixed lanes (lanes with a fixed driving direction) to determine whether the vehicles are violating the lane guidance rules. However, variable lanes (lanes whose driving direction can change dynamically with traffic flow or time) are becoming more and more common on current urban roads. Traffic intersections may include fixed lanes and variable lanes. Solutions using related technologies cannot accurately identify vehicles in variable lanes for violations. Summary of the invention

[0005] The purpose of the embodiments of the present application is to provide a method, device, binocular camera and storage medium for identifying vehicle violations, so as to accurately identify whether a vehicle in a variable lane has violated the regulations. The specific technical solution is as follows:

[0006] In a first aspect, an embodiment of the present application provides a method for identifying vehicle violations, which is applied to a binocular camera, wherein the field of view of a first camera of the binocular camera is a first field of view, and the field of view of a second camera of the binocular camera is a second field of view. The first field of view includes a sign of a variable lane corresponding to a traffic intersection, and a section of the variable lane indicated by the sign, and the second field of view includes the traffic intersection and the lane sections of each lane; the method includes:

[0007] In response to any vehicle entering a road section of a variable lane within the first field of view, the first camera is controlled to collect images, and the collected images are analyzed to obtain a first feature associated with the vehicle; wherein the first feature associated with any vehicle includes a vehicle feature of the vehicle, and a lane number and a sign feature of the variable lane to which the vehicle is traveling; the sign feature is used to represent the driving direction currently indicated in the sign of the variable lane to which the vehicle is traveling;

[0008] In response to any vehicle entering the traffic intersection within the second field of view, controlling the second camera to collect images, and analyzing the collected images to obtain a second feature associated with the vehicle; wherein the second feature associated with any vehicle includes a vehicle feature of the vehicle, a lane number of a lane to which the vehicle is traveling, and a driving direction feature of the vehicle when passing through the traffic intersection;

[0009] For a target vehicle associated with a second feature, based on the vehicle feature of the target vehicle and the vehicle features in each of the first features, a vehicle that is the same as the target vehicle is determined from each of the vehicles associated with the first feature, and the first feature of the determined vehicle is used as a first reference feature;

[0010] In response to the lane number in the first reference feature being the same as the lane number in the second feature associated with the target vehicle, the sign feature in the first reference feature is used as a basis for judgment, and based on the driving direction feature in the second feature associated with the target vehicle, it is judged whether the target vehicle has violated the regulations by not driving according to the guidance.

[0011] In a second aspect, an embodiment of the present application provides a vehicle violation identification device, which is applied to a binocular camera, wherein the field of view of the first camera of the binocular camera is the first field of view, and the field of view of the second camera of the binocular camera is the second field of view. The first field of view includes the sign of the variable lane corresponding to the traffic intersection, and the section of the variable lane indicated by the sign, and the second field of view includes the traffic intersection and the lane sections of each lane; the device includes:

[0012] a first control module, configured to control the first camera to acquire images in response to any vehicle entering a road section of a variable lane within the first field of view, and to analyze the acquired images to obtain a first feature associated with the vehicle; wherein the first feature associated with any vehicle includes a vehicle feature of the vehicle, and a lane number and a sign feature of the variable lane to which the vehicle is traveling; the sign feature is used to represent a driving direction currently indicated by the sign of the variable lane to which the vehicle is traveling;

[0013] a second control module, configured to control the second camera to acquire images in response to any vehicle entering the traffic intersection within the second field of view, and to analyze the acquired images to obtain a second feature associated with the vehicle; wherein the second feature associated with any vehicle includes a vehicle feature of the vehicle, a lane number of a lane to which the vehicle is traveling, and a driving direction feature of the vehicle when passing through the traffic intersection;

[0014] A first determination module is used to determine, for a target vehicle associated with a second feature, a vehicle that is the same vehicle as the target vehicle from among the vehicles associated with the first feature based on the vehicle feature of the target vehicle and the vehicle features in the first features, and use the first feature of the determined vehicle as a first reference feature;

[0015] The first judgment module is used to respond to the lane number in the first reference feature being the same as the lane number in the second feature associated with the target vehicle, take the sign feature in the first reference feature as a judgment basis, and based on the driving direction feature in the second feature associated with the target vehicle, judge whether the target vehicle has violated the regulations by not driving according to the guidance.

[0016] In a third aspect, an embodiment of the present application provides a binocular camera, the binocular camera comprising a first camera, a second camera and a main controller; the field of view of the first camera of the binocular camera is a first field of view, the field of view of the second camera of the binocular camera is a second field of view, the first field of view includes a sign of a variable lane corresponding to a traffic intersection, and a section of the variable lane indicated by the sign, and the second field of view includes the traffic intersection and the lane sections of each lane;

[0017] The first camera is used to collect images of the first field of view under the control of the main controller;

[0018] The second camera is used to collect images of the second field of view under the control of the main controller;

[0019] The main controller is used to control the first camera to collect images of the first field of view in response to any vehicle entering the road section of the variable lane within the first field of view, and analyze the collected images to obtain the first feature associated with the vehicle; in response to any vehicle entering the traffic intersection within the second field of view, control the second camera to collect images of the second field of view, and analyze the collected images to obtain the second feature associated with the vehicle; for a target vehicle associated with the second feature, based on the second vehicle feature of the target vehicle and each first vehicle feature, determine the vehicle that is the same as the target vehicle from each vehicle associated with the first feature, and use the first feature of the determined vehicle as the first reference feature; in response to the first lane number in the first reference feature being the same as the second lane number in the second feature associated with the target vehicle, use the driving direction indicated by the signboard feature in the first reference feature as the judgment basis, and based on the driving direction feature in the second feature associated with the target vehicle, judge whether the target vehicle has violated the rules of not driving according to the guidance;

[0020] Among them, the first feature associated with any vehicle includes: the first vehicle feature of the vehicle, the first lane number of the variable lane in which the vehicle is traveling, and the signboard feature of the variable lane; the signboard feature is used to indicate the driving direction of the variable lane in which the vehicle is traveling; the second feature associated with any vehicle includes: the second vehicle feature of the vehicle, the second lane number of the lane in which the vehicle is traveling, and the driving direction feature of the vehicle.

[0021] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, any of the above-mentioned vehicle violation identification methods is implemented.

[0022] An embodiment of the present application also provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute any of the above-described vehicle violation identification methods.

[0023] Beneficial effects of the embodiments of the present application:

[0024] The vehicle violation identification method provided in the embodiment of the present application is applied to a binocular camera, and the field of view of the binocular camera includes the first field of view of the first camera and the second field of view of the second camera. The first field of view includes the sign of the variable lane and the corresponding road section, and the second field of view includes the traffic intersection and the lane section of each lane; the field of view of the first camera includes the sign of the variable lane, and the first feature associated with any vehicle obtained by analysis includes the vehicle feature of the vehicle, and the lane number and sign feature of the variable lane traveled by the vehicle, and the sign feature is used to characterize the current driving direction indicated in the sign of the variable lane traveled by the vehicle, that is, to associate the various features of the vehicle when entering the variable lane with the vehicle; the field of view of the second camera includes the traffic intersection and each vehicle The second feature of any vehicle obtained by analysis includes the vehicle feature of the vehicle, the lane number of the lane to which the vehicle is traveling, and the driving direction feature when passing through the traffic intersection, that is, the various features of the vehicle when leaving the lane section and passing through the traffic intersection are associated with the vehicle; for the target vehicle associated with the second feature, through the vehicle feature of the target vehicle and the vehicle features in each first feature, the vehicle identical to the target vehicle can be determined from the vehicles associated with the first feature, and the first reference feature of the target vehicle can be obtained; when the lane number in the first reference feature is the same as the lane number in the second feature associated with the target vehicle, the signboard feature in the first reference feature can be used as the basis for judgment, and based on the driving direction feature associated with the target vehicle, it can be accurately judged whether the target vehicle has violated the rules by not following the guidance. It can be seen that the present application uses two cameras to respectively collect images of their corresponding field of view ranges, and can associate various features when the vehicle enters the variable lane with the vehicle, and associate various features when the vehicle leaves the lane section and passes through the traffic intersection with the vehicle, so that after identifying the same target vehicle, the signboard feature when the target vehicle enters the variable lane can be used as the basis for judgment, and the vehicle in the variable lane can be accurately identified whether there is any violation.

[0025] Of course, implementing any product or method of the present application does not necessarily require achieving all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application, and for ordinary technicians in this field, other embodiments can also be obtained based on these drawings.

[0027] Figure 1 A flow chart of a vehicle violation identification method provided in an embodiment of the present application;

[0028] Figure 2 Another schematic diagram of a process for identifying vehicle violation behavior provided by an embodiment of the present application;

[0029] FIG3 (A) is a schematic diagram of the close-range field of view of a binocular camera installed near a traffic intersection according to an embodiment of the present application;

[0030] FIG3 (B) is a schematic diagram of the long-range field of view of a binocular camera installed near a traffic intersection according to an embodiment of the present application;

[0031] Figure 4 Another schematic diagram of a process for identifying vehicle violation behavior provided by an embodiment of the present application;

[0032] Figure 5 A flow chart of a method for identifying whether a vehicle has violated the regulations by not following the guidance provided in an embodiment of the present application;

[0033] Figure 6 A schematic diagram of a scenario for identifying whether a vehicle has violated the regulations by not following the guidance provided in an embodiment of the present application;

[0034] Figure 7 A flow chart of a method for identifying whether a vehicle has violated the lane-changing violation behavior on a solid line provided in an embodiment of the present application;

[0035] Figure 8 A schematic diagram of a scenario for identifying whether a vehicle has violated the lane-changing violation on a solid line provided in an embodiment of the present application;

[0036] Fig. 9 A schematic diagram of an image collected by a vehicle violation identification method provided in an embodiment of the present application;

[0037] Fig.10 A schematic diagram of the structure of a vehicle violation identification device provided in an embodiment of the present application;

[0038] Fig.11 A block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field based on the present application belong to the scope of protection of the present application.

[0040] At present, when identifying vehicle violations, only vehicles in fixed lanes are identified to see if they have violated the lane-changing rule of a solid line. Other violations cannot be identified, and the solution has great limitations. Currently, multiple cameras are usually used to identify vehicles within the range of traffic intersections. There is a large amount of real-time data interaction between multiple cameras, which may seriously affect the efficiency of camera image acquisition. In addition, when the vehicle speed is too fast, the image acquisition between multiple cameras may be out of sync, and the corresponding image may not be acquired.

[0041] However, as variable lanes are becoming more and more common, the driving direction indicated by the signs of the variable lanes may change, making it difficult to identify the violations of vehicles in the variable lanes that do not follow the directions. The recognition accuracy is also low, which is likely to lead to disputes.

[0042] Based on this, the embodiments of the present application provide a vehicle violation identification method, device, binocular camera and storage medium to accurately identify whether a vehicle in a variable lane has violated the regulations.

[0043] The following is an introduction to a vehicle violation identification method provided by this application.

[0044] Among them, a vehicle violation identification method provided in an embodiment of the present application can be applied to a binocular camera, and the binocular camera includes two cameras, namely a first camera and a second camera. In the present application, the field of view of the first camera is the first field of view, and the first field of view can be understood as the field of view including the solid line entry end of the variable lane, which includes the signboard and road section of the variable lane, etc.; the field of view of the second camera is the second field of view, and the second field of view can be understood as the field of view including the solid line exit end of the variable lane, which includes the traffic intersection and the road section of each lane, etc. It can be understood that the variable lane described in the present application can be: a lane whose indication direction can be changed at any time according to the traffic flow, that is, it has no fixed driving direction, and turning or going straight needs to rely on traffic lights or traffic police to change according to the current traffic conditions.

[0045] In addition, a vehicle violation identification method provided in an embodiment of the present application can be applied to any scenario that requires the identification of violations of vehicles in variable lanes, for example: the scenario of identifying violations of vehicles in variable lanes contained in traffic intersections. Of course, traffic intersections can include variable lanes and fixed lanes. The focus of the embodiment of the present application is to identify violations of vehicles in variable lanes, and because the identification method of violations such as solid line lane changes, parking, cutting in, and driving without signal lights of vehicles in variable lanes is similar to the identification method of violations corresponding to vehicles in fixed lanes, therefore, the focus of the present application is to identify violations of vehicles in variable lanes that do not drive according to guidance; of course, the present application can also realize the identification of violations such as solid line lane changes, parking, cutting in, and driving without signal lights of vehicles in variable lanes.

[0046] The execution subject of a vehicle violation identification method provided in an embodiment of the present application can be specifically the main controller of a binocular camera. In this case, the binocular camera can also include a main controller for controlling the first camera and the second camera to acquire images, and analyzing the acquired images. Therefore, the main controller can also be understood as an image processing unit with a function of controlling image acquisition; optionally, the main controller of the binocular camera can also be integrated into other electronic devices outside the binocular camera, such as a mobile phone or a computer, etc., which is not limited in this application. Among them, when the main controller is located in the binocular camera, the main controller can directly obtain the acquisition images of the first camera and the second camera, and when the conditions are met, respectively control the first camera and the second camera to acquire images, and analyze the acquired images; when the main controller is located in an electronic device outside the binocular camera, the electronic device is connected to the binocular camera, and the electronic device can obtain the acquisition images of the first camera and the second camera from the binocular camera, and when the conditions are met, respectively control the first camera and the second camera to acquire images, and obtain the images acquired by the first camera and the second camera from the binocular camera, and analyze the acquired images, etc.

[0047] Specifically, a vehicle violation identification method provided in an embodiment of the present application may include:

[0048] In response to any vehicle entering a road section of a variable lane within the first field of view, the first camera is controlled to collect images within the first field of view, and the collected images are analyzed to obtain a first feature associated with the vehicle; wherein the first feature associated with any vehicle includes: a first vehicle feature of the vehicle, a first lane number of a variable lane on which the vehicle is traveling, and a signboard feature of the variable lane on which the vehicle is traveling; the signboard feature is used to indicate a driving direction of the variable lane on which the vehicle is traveling;

[0049] In response to any vehicle entering the traffic intersection within the second field of view, controlling the second camera to collect images within the second field of view, and analyzing the collected images to obtain a second feature associated with the vehicle; wherein the second feature associated with any vehicle includes: a second vehicle feature of the vehicle, a second lane number of a lane in which the vehicle is traveling, and a feature of a driving direction of the vehicle;

[0050] For a target vehicle associated with a second feature, based on the second vehicle feature of the target vehicle and each first vehicle feature, a vehicle that is the same as the target vehicle is determined from each vehicle associated with the first feature, and the first feature of the determined vehicle is used as a first reference feature;

[0051] In response to the first lane number in the first reference feature being the same as the second lane number in the second feature associated with the target vehicle, the driving direction indicated by the sign feature in the first reference feature is used as a basis for judgment, and based on the driving direction feature in the second feature associated with the target vehicle, it is judged whether the target vehicle has violated the regulations by not driving according to the guidance.

[0052] The vehicle violation identification method provided in the embodiment of the present application is applied to a binocular camera, and the field of view of the binocular camera includes the first field of view of the first camera and the second field of view of the second camera. The first field of view includes the sign of the variable lane and the corresponding road section, and the second field of view includes the traffic intersection and the lane section of each lane; the field of view of the first camera includes the sign of the variable lane, and the first feature associated with any vehicle obtained by analysis includes the vehicle feature of the vehicle, and the lane number and sign feature of the variable lane traveled by the vehicle, and the sign feature is used to characterize the current driving direction indicated in the sign of the variable lane traveled by the vehicle, that is, to associate the various features of the vehicle when entering the variable lane with the vehicle; the field of view of the second camera includes the traffic intersection and each vehicle The second feature of any vehicle obtained by analysis includes the vehicle feature of the vehicle, the lane number of the lane to which the vehicle is traveling, and the driving direction feature when passing through the traffic intersection, that is, the various features of the vehicle when leaving the lane section and passing through the traffic intersection are associated with the vehicle; for the target vehicle associated with the second feature, through the vehicle feature of the target vehicle and the vehicle features in each first feature, the vehicle identical to the target vehicle can be determined from the vehicles associated with the first feature, and the first reference feature of the target vehicle can be obtained; when the lane number in the first reference feature is the same as the lane number in the second feature associated with the target vehicle, the signboard feature in the first reference feature can be used as the basis for judgment, and based on the driving direction feature associated with the target vehicle, it can be accurately judged whether the target vehicle has violated the rules by not following the guidance. It can be seen that the present application uses two cameras to respectively collect images of their corresponding field of view ranges, and can associate various features when the vehicle enters the variable lane with the vehicle, and associate various features when the vehicle leaves the lane section and passes through the traffic intersection with the vehicle, so that after identifying the same target vehicle, the signboard feature when the target vehicle enters the variable lane can be used as the basis for judgment, and the vehicle in the variable lane can be accurately identified whether there is any violation.

[0053] In the following, in conjunction with the accompanying drawings, a vehicle violation identification method provided in an embodiment of the present application is exemplarily introduced.

[0054] like Figure 1 As shown, a vehicle violation identification method provided in an embodiment of the present application is applied to a binocular camera; the method may include the following steps:

[0055] S101: In response to any vehicle entering a lane section of a variable lane within the first field of view, controlling the first camera to collect images, and analyzing the collected images to obtain a first feature associated with the vehicle;

[0056] The first feature associated with any vehicle includes the vehicle feature of the vehicle, and the lane number and sign feature of the variable lane to which the vehicle is traveling; the sign feature is used to represent the driving direction currently indicated in the sign of the variable lane to which the vehicle is traveling;

[0057] In the vehicle violation identification method provided in the embodiment of the present application, a binocular camera is set up near the traffic intersection, and the first field of view of the first camera of the binocular camera includes the sign of the variable lane corresponding to the traffic intersection and the corresponding road section, and the second field of view of the second camera of the binocular camera includes the traffic intersection and the lane sections of each lane. Through the first camera and the second camera of the binocular camera, it is possible to accurately identify whether the vehicle in the variable lane has violated the regulations. Among them, the so-called variable lane corresponding to the traffic intersection can be understood as the variable lane connected to the traffic intersection; and, under the premise of ensuring that the first camera and the second camera of the binocular camera have the corresponding field of view, the specific installation position of the binocular camera and the distance between it and the traffic intersection are not limited by this application.

[0058] For example, FIG3 (A) is a schematic diagram of the near-view field of view of a binocular camera installed near a traffic intersection, and FIG3 (B) is a schematic diagram of the far-view field of view of the binocular camera. The near-view field of view 01 is the first field of view of the first camera, and the near-view field of view 01 includes a sign 02 for a variable lane and a road section 03 of the variable lane indicated by the sign. Of course, the near-view field of view 01 can also include a sign 04 for a fixed lane; the far-view field of view 06 is the second field of view of the second camera, and the far-view field of view 06 includes a traffic intersection and sections of each lane. In addition, as shown in FIG3 (A) and FIG3 (B), there are also signs (signs 04 for fixed lanes and signs 02 for variable lanes) and traffic lights 05 at the traffic intersection. It should be noted that the black arrow in the sign 02 for the variable lane is the current driving direction indicated by the variable lane, and the gray arrow is the variable driving direction of the variable lane.

[0059] First, when any vehicle enters a road section of a variable lane within the first field of view, the first camera is controlled to collect images, and the collected images are analyzed to obtain the first feature associated with the vehicle. Among them, the first feature associated with any vehicle includes the vehicle feature of the vehicle, as well as the lane number and signboard feature of the variable lane where the vehicle travels. The vehicle feature in the first feature is used to match the vehicle feature in the subsequent second feature, the lane number of the variable lane is used to match the lane number in the subsequent second feature, and the signboard feature is used to characterize the current driving direction indicated when the vehicle travels to the signboard of the variable lane, which is used as a subsequent judgment basis to accurately identify whether the vehicle in the variable lane has violated the regulations.

[0060] In addition, when the first camera is controlled to collect images, the collected images include the vehicle, the variable lane where the vehicle is traveling, and the sign of the variable lane where the vehicle is traveling. By performing object recognition and other analyses on the collected images, the first feature associated with the vehicle can be obtained. The specific implementation method is similar to the prior art and will not be repeated here. In the present application, a binocular camera is set up at the traffic intersection to accurately identify whether the vehicle in the variable lane has violated the regulations. Therefore, the first field of view may be relatively large, for example: including the sign of each lane corresponding to the traffic intersection, and the road section corresponding to each lane. At this time, the image obtained by controlling the first camera to collect images may include other lanes, vehicles in other lanes, and signs of other lanes; other lanes may be variable lanes and / or fixed lanes, wherein the sign of the fixed lane indicates a fixed driving direction, and the driving direction indicated by the sign of the variable lane may change with time or traffic volume, and the present application does not limit this.

[0061] It should be noted that in this application, the captured image of the first camera can be used to identify whether any vehicle has entered the lane section of the variable lane within the first field of view; for example: for the captured image of the first camera, the vehicle area intrusion detection is performed in real time to detect whether the vehicle has entered the lane section of the variable lane within the first field of view; wherein, the area within the first field of view can be configured, for example: the first field of view is configured as a dotted line driving area and a solid line driving area, and the solid line driving area can also be divided into a variable lane driving area, or also includes a fixed lane driving area; when a vehicle is detected to enter the variable lane driving area, the vehicle is identified to enter the lane section of the variable lane within the first field of view. Of course, any method that can identify the vehicle entering the lane section of the variable lane within the first field of view is applicable to this application, and this application does not limit this.

[0062] S102: In response to any vehicle entering the traffic intersection within the second field of view, controlling the second camera to collect images, and analyzing the collected images to obtain a second feature associated with the vehicle;

[0063] The second feature associated with any vehicle includes the vehicle feature of the vehicle, the lane number of the lane to which the vehicle is traveling, and the driving direction feature of the vehicle when passing through the traffic intersection;

[0064] After entering the first field of view, the vehicle will enter the second field of view after driving on the road section of the variable lane. The second field of view includes the traffic intersection and the road sections of each lane. Through the driving behavior of the vehicle in the second field of view and the first feature obtained by analyzing the vehicle in the first field of view, it is possible to accurately identify whether the vehicle in the variable lane has any illegal behavior; when any vehicle enters the second field of view, the second camera can be controlled to collect images, and the collected images can be analyzed to obtain the second feature associated with the vehicle. Among them, the second feature associated with any vehicle includes the vehicle feature of the vehicle, the lane number of the lane to which the vehicle is driving, and the driving direction feature of the vehicle when passing through the traffic intersection. The vehicle feature in the second feature is used to match the vehicle feature in the first feature, so as to identify the vehicle's illegal behavior based on the first and second features associated with the same vehicle; the lane number in the second feature is used to match the lane number of the variable lane in the first feature, and the driving direction feature in the second feature represents the driving direction of the vehicle when passing through the traffic intersection.

[0065] When the second camera is controlled to capture images, the captured images include the vehicle, the lane the vehicle is traveling to, and the direction of travel of the vehicle. Object recognition and other analyses of the captured images can obtain a second feature associated with the vehicle. The specific implementation method is similar to the analysis method of the first feature and will not be repeated here. The first field of view range can be understood as the near-field of view range of the binocular camera, and the second field of view range can be understood as the far-field of view range of the binocular camera. The second field of view range is larger than the first field of view range. At this time, the image obtained by controlling the second camera for image capture may include: roads and vehicles passing through the traffic intersection in other directions, etc. This application does not limit this.

[0066] The so-called entering the traffic intersection within the second field of view refers to leaving the section of the variable lane and entering the traffic intersection. In the present application, the capture image of the second camera can be used to identify whether any vehicle has entered the traffic intersection within the second field of view. For example, based on the capture image of the second camera, real-time vehicle area intrusion detection is performed to detect whether the vehicle has entered the traffic intersection within the second field of view. Among them, the area within the second field of view can be configured, for example, the second field of view is configured as a traffic intersection area and a lane area to be entered into the traffic intersection. When a vehicle is detected to enter the traffic intersection area from the lane area, the vehicle is identified as entering the traffic intersection within the second field of view.

[0067] It should be noted that the first field of view range can be called the near-field field of view range, which is mainly used to identify the status of the vehicle when it enters the variable lane, such as: the characteristics of the vehicle itself, the lane number of the variable lane in which the vehicle is traveling, the driving direction indicated by the sign of the variable lane, etc.; the second field of view range can be called the long-field field of view range, which is mainly used to identify the status of the vehicle when it leaves the lane section and enters the traffic intersection, such as: the characteristics of the vehicle itself, the lane number in which the vehicle was traveling before entering the traffic intersection, and the driving direction of the vehicle when entering the traffic intersection, etc.; the status of the vehicle recorded by the first field of view range and the second field of view range can be used to accurately identify the violations of vehicles in the variable lane through subsequent steps.

[0068] It should be emphasized that the present application can also control the second camera to capture images (i.e., the subsequent long-range image) and analyze the captured images when any vehicle enters the second field of view. At this time, only the vehicle features of the vehicle can be analyzed for matching with the vehicle features in the first features. After the match is successful, the driving behavior of the vehicle is identified in real time, the second camera is controlled to capture images (i.e., the subsequent illegal driving image), and the captured images are analyzed. At this time, the lane number and driving direction characteristics of the vehicle are analyzed for subsequent judgment on whether the vehicle has any illegal behavior.

[0069] S103: for a target vehicle associated with a second feature, based on the vehicle feature of the target vehicle and the vehicle features in each first feature, determine a vehicle that is the same as the target vehicle from each vehicle associated with the first feature, and use the first feature of the determined vehicle as a first reference feature;

[0070] After obtaining the first feature associated with the vehicle and the second feature associated with the vehicle, the vehicle violation can be identified. The identification of violation is to identify whether the second feature has a violation. The first feature can be used as a basis for judging whether the second feature has a violation. First, the first reference feature corresponding to the second feature can be determined. Specifically, for the target vehicle associated with the second feature, based on the vehicle features of the target vehicle and the vehicle features in each first feature, the vehicle that is the same as the target vehicle is determined from the vehicles associated with the first feature. The first feature of the determined vehicle can be used as a basis for judging the second feature of the target vehicle, and the first feature of the determined vehicle is used as the first reference feature. Among them, any vehicle associated with the second feature can be used as a target vehicle.

[0071] It is understandable that the vehicle feature of any vehicle may only include one-dimensional features, such as the license plate number; of course, the vehicle feature of any vehicle may include multi-dimensional features, such as the license plate number and other features, such as the vehicle model, vehicle body color, etc. This application does not limit the number of feature dimensions included in the vehicle feature.

[0072] Exemplarily, in the case where the vehicle feature of any vehicle includes one-dimensional or multi-dimensional features, based on the vehicle feature of the target vehicle and the vehicle features in each first feature, determining a vehicle that is the same vehicle as the target vehicle from each vehicle associated with the first feature may include:

[0073] The license plate number of the target vehicle is matched with the license plate numbers in each first feature, and a vehicle whose license plate number matches successfully among the vehicles associated with the first feature is determined to obtain a vehicle that is the same vehicle as the target vehicle.

[0074] Exemplarily, when the vehicle features of any vehicle include multi-dimensional features, in order to more accurately identify, the features of other dimensions may be used as auxiliary features while using the license plate number. Accordingly, based on the vehicle features of the target vehicle and the vehicle features in each first feature, determining a vehicle that is the same vehicle as the target vehicle from each vehicle associated with the first feature may include:

[0075] The vehicle features of the target vehicle are matched with the vehicle features in each first feature, and the vehicle whose license plate number and at least one target dimension feature are successfully matched in each vehicle associated with the first feature are determined, and the vehicle that is the same vehicle as the target vehicle is obtained. The at least one target dimension can be at least one dimension of the vehicle features other than the license plate number.

[0076] S104: In response to the lane number in the first reference feature being the same as the lane number in the second feature associated with the target vehicle, taking the sign feature in the first reference feature as a basis for judgment and based on the driving direction feature in the second feature associated with the target vehicle, judging whether the target vehicle has violated the regulations by not driving according to the guidance.

[0077] After determining the target vehicle is the same as the target vehicle, the first reference feature and the second feature of the target vehicle can be used to identify whether the target vehicle has violated the rules. In this application, it is mainly aimed at the target vehicle in the variable lane (when entering the road section within the first field of view, the vehicle is in a variable lane) to identify whether the target vehicle has violated the rules by not following the guidance. Among them, the sign feature in the first reference feature represents the driving direction indicated when the target vehicle drives into the sign of the variable lane. The sign feature in the first reference feature can be used as a judgment basis to identify whether the driving direction feature in the second feature associated with the target vehicle has violated the rules by not following the guidance. At this time, it can be accurately judged whether the target vehicle has violated the rules by not following the guidance.

[0078] Exemplarily, in one implementation, the judging method includes taking the sign feature in the first reference feature as a basis for judging and judging whether the target vehicle has violated the rules of not following the guidance based on the driving direction feature in the second feature associated with the target vehicle:

[0079] If the direction represented by the driving direction feature in the second feature associated with the target vehicle is different from the driving direction indicated by the sign feature in the first reference feature, it is determined that the target vehicle has violated the regulations by not driving according to the guidance.

[0080] Under normal circumstances, the target vehicle will not change lanes and should drive in the direction indicated by the sign when entering the changeable lane. If the direction represented by the driving direction feature in the second feature associated with the target vehicle is different from the driving direction indicated by the sign feature in the first reference feature, the target vehicle does not drive according to the guidance, that is, the target vehicle has violated the regulations by not driving according to the guidance.

[0081] In addition, in some cases, after a vehicle enters a road section with a variable lane in the second field of view, the driving direction indicated by the sign of the variable lane may change, or the sign of the variable lane may malfunction. In this case, when the vehicle enters a traffic intersection in the second field of view, the driving direction indicated by the sign of the variable lane is different from the driving direction represented by the sign characteristics. For example, when a vehicle enters a road section with a variable lane in the first field of view, the driving direction currently indicated by the sign of the variable lane to which the vehicle is traveling is a right turn, that is, the driving direction represented by the sign characteristics is a right turn, and the vehicle enters After the road section with variable lanes, the predetermined time for the direction switching of the sign is reached, for example, at 19:00, the driving direction indicated by the sign is switched to straight ahead, that is, at the moment when the vehicle enters the traffic intersection within the second field of view, the driving direction indicated by the sign of the variable lane is straight ahead, but the driving direction indicated by the sign feature when the vehicle enters the road section with variable lanes within the first field of view is a right turn. At this time, the vehicle should turn right according to the driving direction indicated by the sign of the variable lane when entering the first field of view. If the vehicle drives in the straight ahead direction, the vehicle has violated the regulations by not following the guidance. It can be seen that in the case where the direction of the sign changes or the sign fails, the present application can associate the sign feature when the vehicle enters the variable lane, and use this as the basis for judgment. When the lane number in the first reference feature is the same as the lane number in the second feature associated with the target vehicle, it can accurately judge whether the target vehicle has violated the regulations by not following the guidance.

[0082] In the technical solution of the present application, the operations of acquiring, storing, using, processing, transmitting, providing and disclosing the first feature, second feature and other features of the vehicle involved are all carried out with the user's authorization.

[0083] The vehicle violation identification method provided in the embodiment of the present application is applied to a binocular camera, and the field of view of the binocular camera includes the first field of view of the first camera and the second field of view of the second camera. The first field of view includes the sign of the variable lane and the corresponding road section, and the second field of view includes the traffic intersection and the lane section of each lane; the field of view of the first camera includes the sign of the variable lane, and the first feature associated with any vehicle obtained by analysis includes the vehicle feature of the vehicle, and the lane number and sign feature of the variable lane traveled by the vehicle, and the sign feature is used to characterize the current driving direction indicated in the sign of the variable lane traveled by the vehicle, that is, to associate the various features of the vehicle when entering the variable lane with the vehicle; the field of view of the second camera includes the traffic intersection and each vehicle The second feature of any vehicle obtained by analysis includes the vehicle feature of the vehicle, the lane number of the lane to which the vehicle is traveling, and the driving direction feature when passing through the traffic intersection, that is, the various features of the vehicle when leaving the lane section and passing through the traffic intersection are associated with the vehicle; for the target vehicle associated with the second feature, through the vehicle feature of the target vehicle and the vehicle features in each first feature, the vehicle identical to the target vehicle can be determined from the vehicles associated with the first feature, and the first reference feature of the target vehicle can be obtained; when the lane number in the first reference feature is the same as the lane number in the second feature associated with the target vehicle, the signboard feature in the first reference feature can be used as the basis for judgment, and based on the driving direction feature associated with the target vehicle, it can be accurately judged whether the target vehicle has violated the rules by not following the guidance. It can be seen that the present application uses two cameras to respectively collect images of their corresponding field of view ranges, and can associate various features when the vehicle enters the variable lane with the vehicle, and associate various features when the vehicle leaves the lane section and passes through the traffic intersection with the vehicle, so that after identifying the same target vehicle, the signboard feature when the target vehicle enters the variable lane can be used as the basis for judgment, and the vehicle in the variable lane can be accurately identified whether there is any violation.

[0084] The vehicle violation identification method provided in the embodiment of the present application can also identify, for a target vehicle, whether the target vehicle has other violations other than not driving according to the guidance. Optionally, the method further includes:

[0085] In response to the lane number in the first reference feature being different from the lane number in the second feature associated with the target vehicle, it is determined that the target vehicle has violated the lane change rule by crossing a solid line.

[0086] It can be understood that after the vehicle enters the road section with a variable lane within the first field of view, the road sections at this time are all solid line driving sections. If the lane number in the first reference feature is different from the lane number in the second feature associated with the target vehicle, it means that a solid line lane change violation has occurred in the variable lane where the target vehicle is traveling. At this time, the difference between the two lane numbers analyzed under the two fields of view can be used to determine that the target vehicle has a solid line lane change violation.

[0087] Optionally, the first feature associated with any vehicle further includes: the driving order of the vehicle in the variable lane to which it is traveling; the second feature associated with any vehicle further includes: the driving order of the vehicle in the lane to which it is traveling;

[0088] The method further comprises:

[0089] In response to the lane number in the first reference feature being the same as the lane number in the second feature associated with the target vehicle, it is determined whether the target vehicle has violated the lane change rule by changing lanes along a solid line based on the driving sequence in the first reference feature and the driving sequence in the second feature associated with the target vehicle.

[0090] When analyzing the first feature and the second feature associated with the vehicle, the driving order of the lane to which the vehicle is traveling can also be analyzed. At this time, the first feature associated with any vehicle also includes the driving order of the vehicle in the variable lane, and the second feature associated with any vehicle also includes the driving order of the vehicle in the lane; since the lane sections after the vehicle enters the lane section of the variable lane within the first field of view are all solid line driving sections, at this time, it is also possible to judge whether the vehicle has a solid line lane change violation by the driving order of the vehicle; illustratively: when the lane number in the first reference feature is the same as the lane number in the second feature associated with the target vehicle, but the driving order in the first reference feature is different from the driving order in the second feature associated with the target vehicle, it means that the target vehicle may have a solid line lane change violation, and at this time it is necessary to further accurately judge whether the target vehicle has a solid line lane change violation; if the driving order in the first reference feature is the same as the driving order in the second feature associated with the target vehicle, it means that the target vehicle does not have a solid line lane change violation.

[0091] Specifically, if the driving order in the first reference feature indicates that the target vehicle is in the fifth position, and the driving order in the second feature associated with the target vehicle indicates that the target vehicle is in the fourth position, then the target vehicle has changed lanes and overtaken the previous vehicle on the solid line driving section, and the target vehicle has violated the solid line lane change rule, or the driving lane of a vehicle before the target vehicle has changed, and the target vehicle has not violated the solid line lane change rule. If the driving order in the first reference feature indicates that the target vehicle is in the fifth position, and the driving order in the second feature associated with the target vehicle indicates that the target vehicle is in the sixth position, then it means that the target vehicle has been overtaken by other vehicles on the solid line driving section, and the target vehicle has not violated the solid line lane change rule.

[0092] Optionally, the first feature associated with any vehicle further includes a first time, wherein the first time is a time for performing image acquisition in response to the vehicle entering a lane section of a variable lane within the first field of view;

[0093] The second feature associated with any vehicle further includes a second time, wherein the second time is a time for performing image acquisition in response to the vehicle entering a traffic intersection within the second field of view;

[0094] The method further comprises:

[0095] In response to the lane number in the first reference feature being the same as the lane number in the second feature associated with the target vehicle, determining whether the target vehicle has committed a parking violation based on a first time in the first reference feature and a second time in the second feature associated with the target vehicle;

[0096] In response to the lane number in the first reference feature being different from the lane number in the second feature associated with the target vehicle, it is determined whether the target vehicle has committed a lane-cutting violation based on a first time in the first reference feature and a second time in the second feature associated with the target vehicle.

[0097] When the first feature and the second feature associated with the vehicle are analyzed, the first time and the second time of image acquisition can also be recorded respectively. At this time, the first feature associated with any vehicle also includes the first time of image acquisition when the vehicle enters the variable lane, and the second feature associated with any vehicle also includes the second time of image acquisition when the vehicle enters the traffic intersection. Since the lane sections after the vehicle enters the lane section of the variable lane within the first field of view are all solid line driving sections, and the driving section is located near the traffic intersection, there are also driving time requirements for the driving vehicles. At this time, the first time and the second time of the target vehicle can be used to determine whether the target vehicle has any parking or lane-cutting violations. Exemplary: when the lane number in the first reference feature is the same as the lane number in the second feature associated with the target vehicle, but the time difference between the first time and the second time is greater than the first driving time (such as 60s, etc.), that is, the target vehicle stays on the road section for a long time, but does not change lanes, then it means that the target vehicle has a parking violation; when the lane number in the first reference feature is different from the lane number in the second feature associated with the target vehicle, if the time difference between the first time and the second time is greater than the second driving time, that is, the target vehicle stays on the road section for a long time and changes lanes, then it means that the target vehicle has a blocking violation. Among them, the first driving time and the second driving time are both the maximum allowable driving time of the road section of each lane, the first driving time is the recognition threshold for parking violations, and the second driving time is the recognition threshold for blocking violations. The first driving time and the second driving time can be the same or different, and this application does not limit this.

[0098] It should be noted that the first feature associated with any vehicle may include the driving order and the first time at the same time, and the second feature associated with any vehicle may include the driving order and the second time at the same time; or, the first feature associated with any vehicle includes the driving order of the variable lanes, and the second feature associated with any vehicle includes the driving order of the lanes; or, the first feature associated with any vehicle includes the first time, and the second feature associated with any vehicle includes the second time. That is, the first feature associated with any vehicle and the associated second feature contain the same dimensions regarding the driving order and the time when the image was captured, and may both contain the dimensions of the driving order and the time, or both only contain the dimension of the driving order, or both only contain the dimension of the time when the image was captured, and this application does not limit this.

[0099] The vehicle violation identification method provided in the embodiment of the present application can also identify whether the vehicle has violated the traffic rules by not following traffic lights, and the second feature associated with any vehicle also includes: when the vehicle passes through the traffic intersection, the traffic light feature of the traffic intersection;

[0100] The method further comprises:

[0101] Based on the traffic light feature in the second feature associated with the target vehicle, it is determined whether the target vehicle has a violation of driving without following traffic lights.

[0102] When any vehicle enters the traffic intersection within the second field of view, the second camera may be controlled to collect images including traffic lights. At this time, the traffic lights may be analyzed to obtain traffic light features at the traffic intersection. The traffic light features are used to characterize whether the road sections of each lane are allowed to pass. At this time, for the target vehicle, the traffic light features in the second feature may be used to determine whether the target vehicle has violated the regulations by not following the traffic lights.

[0103] Specifically, judging whether the target vehicle has violated the traffic rules by not following traffic lights based on the traffic light feature in the second feature associated with the target vehicle includes:

[0104] In response to determining that the target vehicle does not have a violation of driving without following the guidance, determining whether the target vehicle has a violation of driving without following the signal light based on a first signal light feature; wherein the first signal light feature is a signal light feature about a first signal light in a second feature associated with the target vehicle, and the first signal light is a signal light corresponding to the driving direction indicated by the sign feature in the first reference feature;

[0105] In response to determining that the target vehicle has violated the regulations by not driving according to the guidance, a determination is made based on the second signal light feature whether the target vehicle has violated the regulations by not driving according to the signal light; wherein the second signal light feature is: the signal light feature for the second signal light in the second feature associated with the target vehicle, and the second signal light is: the signal light corresponding to the driving direction represented by the driving direction feature in the second feature associated with the target vehicle.

[0106] When determining whether the target vehicle has committed the violation of not following traffic lights, it is possible to first determine whether the target vehicle has committed the violation of not following directions, and whether there are consequences of not following directions, which will affect the basis for determining whether the target vehicle has committed the violation of not following traffic lights.

[0107] Specifically, if the target vehicle does not violate the regulations by not following the guidance, the first signal light feature corresponding to the driving direction indicated by the signboard feature can be used as a basis for judgment to determine whether the target vehicle has violated the regulations by not following the signal light; if the target vehicle has violated the regulations by not following the guidance, the second signal light feature corresponding to the driving direction represented by the driving direction feature can be used as a basis for judgment to determine whether the target vehicle has violated the regulations by not following the signal light. Wherein, when the target vehicle does not violate the regulations by not following the guidance, the driving direction feature of the target vehicle is the same as the driving direction represented by the signboard feature. At this time, the first signal light and the second signal light are the same signal light. The first signal light feature or the second signal light feature can be used as a basis for judgment to determine whether the target vehicle has violated the regulations by not following the signal light. When the target vehicle violates the regulations by not following the guidance, the driving direction feature of the target vehicle is different from the driving direction represented by the signboard feature. At this time, the second signal light feature of the second signal light corresponding to the driving direction feature can be used as a basis for judgment to determine whether the target vehicle has violated the regulations by not following the signal light.

[0108] In addition, the target vehicle should actually follow the driving direction indicated by the signboard characteristics. Regardless of whether the target vehicle has violated the regulations by not driving according to the guidance, the first signal light characteristics of the first signal light corresponding to the driving direction indicated by the signboard characteristics can be used as a basis for judgment to identify whether the target vehicle has violated the regulations by not driving according to the signal light. This is reasonable.

[0109] It can be seen that the vehicle violation identification method provided in the embodiment of the present application may further include other features in the first feature associated with the vehicle, and may further include other features in the second feature associated with the vehicle. It can be identified whether the target vehicle has committed any violation such as changing lanes with a solid line, stopping, cutting in, or driving without following traffic lights based on the first reference feature and the second feature associated with the target vehicle.

[0110] Optionally, in another embodiment of the present application, Figure 2 As shown, the method further comprises the following steps:

[0111] S201: In response to any vehicle entering a fixed lane within the first field of view, controlling the first camera to collect images, and analyzing the collected images to obtain a third feature associated with the vehicle;

[0112] The third feature associated with any vehicle includes the vehicle feature of the vehicle, and the lane number and lane travel direction of the fixed lane traveled by the vehicle;

[0113] In the present application, the first field of view also includes a fixed lane and a road section of the fixed lane. When any vehicle enters the fixed lane, the first camera can be controlled to capture images and analyze the captured images to obtain a third feature associated with the vehicle.

[0114] It should be noted that step S201 and step S101 can be performed simultaneously, that is, within the first field of view, there is a vehicle entering a road section with a variable lane, and at the same time there is another vehicle entering a fixed lane. At this time, an image is acquired and analyzed to obtain the first feature of the vehicle in the variable lane and the third feature of the vehicle in the fixed lane. In addition, the analysis method for obtaining the third feature is similar to the analysis method for obtaining the first feature. When obtaining the third feature, it is not necessary to analyze the signboard feature, but the direction feature of the fixed lane can be analyzed, which will not be described in detail here.

[0115] S202: for a designated vehicle associated with the second feature, based on the vehicle feature of the designated vehicle and the vehicle features in each third feature, determine a vehicle that is the same vehicle as the designated vehicle from among the vehicles associated with the third feature, and use the third feature of the determined vehicle as a second reference feature;

[0116] When identifying the violation of a designated vehicle, the vehicle that is the same as the designated vehicle can be first determined from the vehicles associated with the third feature, and the third feature of the determined vehicle can be used as the second reference feature. The specific implementation method is similar to step S103 and will not be repeated here.

[0117] S203: Taking the lane driving direction in the second reference feature as a judgment basis, based on the lane number in the second reference feature, and the lane number and driving direction features in the second feature of the designated vehicle, judging whether the designated vehicle has violated the rules.

[0118] After determining the vehicle that is the same as the designated vehicle, the second reference feature and the second feature of the designated vehicle can be used to identify whether the designated vehicle has violated the rules; among which, violations such as changing lanes on a solid line, parking, cutting in, and driving without following traffic lights are similar to the above-mentioned embodiments and are not described in detail here. When identifying whether the designated vehicle has violated the rules by not following the guidance, the direction feature of the fixed lane and the driving direction feature in the second feature associated with the designated vehicle can be judged to be the same to determine whether the designated vehicle has violated the rules by not following the guidance.

[0119] The vehicle violation identification method provided in the embodiment of the present application can control the first camera to collect images for vehicles in fixed lanes, analyze the collected images to obtain the third feature associated with the vehicle, and use a similar method to the above to identify whether the vehicle in the fixed lane has violated the rules. It can be seen that the present application uses two cameras to collect images in their corresponding fields of view, associates each feature when the vehicle enters the fixed lane with the vehicle, and associates each feature when the vehicle leaves the lane section and passes through the traffic intersection with the vehicle. After identifying the same designated vehicle, the lane driving direction in the second reference feature is used as the judgment basis to accurately identify whether the vehicle in the fixed lane has violated the rules.

[0120] A vehicle violation identification method provided by the present application is introduced below in conjunction with a specific embodiment.

[0121] At present, variable lanes and tidal lanes are becoming more and more common in urban roads. Variable lanes can flexibly adjust the traffic direction of the lanes, improve road utilization, and reduce the occurrence of congestion. The allowed driving direction of variable lanes will change dynamically with traffic volume or time; this application can adapt to the variability of variable lanes and accurately identify whether vehicles in variable lanes have violated regulations.

[0122] In the embodiment of the present application, the scene setup diagram of the binocular camera is shown in Figure 3 (A) and Figure 3 (B). The field of view of the first camera of the binocular camera is the near-view field of view range 01 (i.e., the first field of view range), and the field of view of the second camera of the binocular camera is the far-view field of view range 06 (i.e., the second field of view range). The near-view field of view includes the variable lane sign 02, the fixed lane sign 04, and the road section portion indicated by each lane (including the variable lane section portion 03 and the fixed lane section portion), which can be understood as the area range of the entrance end including the solid line area; the far-view field of view range 06 includes the traffic intersection, the lane sections of each lane, and the traffic light 05, etc., which can be understood as the area range of the exit end including the solid line area.

[0123] Among them, when the vehicle enters the near-view field of view, the binocular camera controls the first camera to collect images, analyzes the collected images, identifies the vehicle features such as the vehicle license plate number, vehicle model, body color, and the lane number of the variable lane in the variable lane area and the driving direction indicated by the sign of the variable lane (i.e., the first feature), which are used to match with the distant field of view. When the vehicle enters the distant field of view, the binocular camera controls the second camera to collect images, analyzes the collected images, identifies the vehicle features such as the vehicle license plate number, vehicle model, body color, and the lane number and driving direction of the vehicle (i.e., the second feature), and matches them with the results of the near-view field of view. After the match is successfully determined to be the same vehicle, the vehicle is subsequently judged whether there is any violation.

[0124] The logic of the above process is as follows Figure 4 As shown, the following steps may be included:

[0125] S401: Recognizing that a vehicle enters the near-view field of view: that is, recognizing that a vehicle enters the near-view field of view through the first camera.

[0126] S402: Identify the vehicle license plate number, lane number, sign direction, and vehicle model, generate a vehicle ID, and record it in a cache queue;

[0127] That is, when a vehicle enters the close-range field of view, the first camera is controlled to capture images, and the captured images are analyzed to identify vehicle features such as the vehicle license plate number, vehicle model, body color, and close-range data such as the lane number of the variable lane in the variable lane area entered and the driving direction indicated by the sign of the variable lane, and generate a vehicle ID, and the identified close-range data is associated with the vehicle and stored in the cache queue.

[0128] S403: recognizing that a vehicle enters the long-range field of view; that is, recognizing that a vehicle enters the long-range field of view through the second camera.

[0129] S404: Identify the vehicle license plate number, vehicle model, etc., and match them with the close-up data in the cache queue;

[0130] When a vehicle enters the long-range field of view, the binocular camera controls the second camera to capture images and analyzes the captured images to identify vehicle features such as the vehicle license plate number, vehicle model, body color, and long-range data such as the vehicle's lane number and direction of travel (or only identify features such as the license plate number, vehicle model, and body color), and match the long-range data of the identified target vehicle with the various close-range data in the cache queue to determine the vehicle that is the same as the target vehicle and obtain the close-range data associated with the target vehicle.

[0131] S405: Identify whether the vehicle is in a variable lane area when entering the near-view field of view;

[0132] That is, it is identified whether the lane in which the target vehicle is traveling when entering the near-view field of view belongs to the variable lane area. If yes, step S406 is executed; if not, step S407 is executed.

[0133] S406: Using the lane direction of the sign matched in the cache queue as a basis for judgment;

[0134] That is, the lane in which the target vehicle is traveling when it enters the near-field of view belongs to the variable lane area, and the lane direction of the sign matched in the cache queue should be used as the basis for subsequent violation judgments.

[0135] S407: Using the direction of the matched fixed lane in the cache queue as a basis for judgment;

[0136] That is, when the target vehicle enters the near-field field of view, the lane it is traveling in does not belong to the variable lane area but to the fixed lane area. At this time, the direction of the fixed lane matched in the cache queue should be used as the basis for subsequent violation judgment.

[0137] S408: Determine whether there is any violation;

[0138] After determining the judgment basis, the judgment basis and the near-view data and long-view data associated with the vehicle can be combined to determine whether the vehicle has violated the rules. The specific judgment method can be referred to the above embodiment and will not be described in detail here. Alternatively, if the long-view data only contains features such as the license plate number, vehicle model, and body color, the vehicle's driving behavior needs to be further identified, for example, the lane number, driving direction, etc., and combined with the judgment basis, etc., to determine whether the vehicle has violated the rules.

[0139] It can be seen that the present application uses two cameras to respectively collect images of their corresponding fields of view, and can associate various features of a vehicle when it enters a lane (a variable lane or a fixed lane) with the vehicle, and also associate various features of the vehicle when it leaves a lane section and passes through a traffic intersection with the vehicle. After identifying the same target vehicle, the corresponding judgment basis is determined according to whether the lane in which the target vehicle enters the near-field of view is a variable lane, and the vehicles in the variable lane and the fixed lane can be accurately identified as to whether they have committed any violations.

[0140] For example, in one implementation, Figure 5 As shown, the method for identifying whether a vehicle has violated the regulations by not driving according to the guidance includes the following steps:

[0141] S501: A vehicle appears in the near-field field of view, and an image is collected to obtain the direction of the sign when the vehicle enters the variable lane;

[0142] S502: The vehicle appears in the distant field of view, and the driving direction is inconsistent with the direction of the sign;

[0143] S503: Determine that the vehicle violates the regulations by not following the guidance.

[0144] like Figure 6 As shown, when the vehicle enters the variable lane within the near-field vision range, the corresponding sign direction is a left turn, and the vehicle goes straight forward and gradually enters the traffic intersection. After entering the traffic intersection within the long-field vision range, the vehicle's driving direction is still straight. At this time, the driving direction is inconsistent with the direction of the sign when entering the variable lane. It can be directly judged that the vehicle has violated the regulations by not driving according to the guidance.

[0145] For example, in one implementation, Figure 7 As shown, the method for identifying whether a vehicle has violated the lane changing rules on a solid line includes the following steps:

[0146] S701: The vehicle appears in the near field of view and the lane number is recorded;

[0147] S702: The vehicle appears in the distant field of view and the lane number changes;

[0148] S703: Determine that the vehicle has violated the lane change rules by changing lanes on a solid line.

[0149] like Figure 8 As shown, when entering the near-field of view, vehicle A is driving in lane 1 and vehicle B is driving in lane 2; when entering the far-field of view, vehicle A is driving in lane 2 and vehicle B is driving in lane 1. The lane numbers of vehicle A and vehicle B have changed. At this time, it is determined that both vehicle A and vehicle B have violated the solid line lane change violation.

[0150] Exemplarily, in one implementation, when the lane number of the vehicle entering the long-distance viewing field from the near-view field of view does not change, and the time difference of the vehicle entering the long-distance viewing field from the near-view field of view exceeds the first driving time, it can be determined that the vehicle has violated parking regulations.

[0151] Exemplarily, in one implementation, when the vehicle is in the far-field field of view, it is possible to identify whether the vehicle has violated the rules by not following the signal lights based on the signal lights corresponding to the direction of the variable lane sign identified by the vehicle in the near-field field of view. At this time, the basis for identifying whether the vehicle has not followed the signal lights is the signal lights corresponding to the direction of the variable lane sign when the vehicle enters the near-field field of view (the second signal lights), rather than the signal lights corresponding to the driving direction of the vehicle in the far-field field of view (the first signal lights).

[0152] Exemplarily, in one implementation, when the lane number of a vehicle entering a long-distance viewing field from a near-view viewing field changes, and the time difference for the vehicle to enter a long-distance viewing field from a near-view viewing field exceeds a second driving time, it can be determined that the vehicle has violated the lane-squeezing rule.

[0153] A schematic diagram of an image collected by a vehicle violation identification method provided in this application is as follows Fig. 9 As shown, the present application uses a binocular camera to capture images of vehicles. First, the vehicle enters a variable lane within a near-view field of view, and images are captured within the near-view field of view to obtain a near-view image, and the vehicle license plate information and the direction of the sign of the variable lane are identified; then, the vehicle enters a far-view field of view, and images are captured within the far-view field of view to obtain a far-view image, and the vehicle license plate information, etc. are identified and matched with the near-view image, and the same vehicle is identified by the license plate number; finally, the vehicle passes through a traffic intersection, and it is identified whether the vehicle with the license plate number has any violations. When there are violations, image capture is performed to obtain an illegal driving image.

[0154] The present application provides a method for identifying vehicle violations, which is mainly for vehicles in variable lanes, and additionally adds the collection of close-up images, analyzes the close-up images to obtain the license plate number and the direction of the sign of the variable lane where the vehicle is traveling, and associates the sign when the vehicle enters the variable lane with the vehicle in reverse, so as to serve as the basis for judging the subsequent violation identification. At present, the identification of violations of vehicles in variable lanes usually adopts fixed-time or timed image acquisition, and there is a time gap, and it is impossible to determine the direction of the sign when the vehicle enters the variable lane when it has entered the long-range field of view at the time of switching the direction of the adjacent sign; for example: the switching time of the sign direction is 19:00, and the scheme using the relevant technology cannot accurately judge the violation of the vehicle that enters the variable lane before 19:00 but does not leave the variable lane after 19:00 (that is, the vehicle that is already in the variable lane and the vehicle that enters the variable lane at the time of switching, etc.); thus, the identification of the violation of the vehicle in the variable lane lacks a basis for judgment, and leads to great controversy and lack of persuasiveness for the judgment result. When a vehicle enters a variable lane, this application associates the vehicle with the direction of the sign in the variable lane, and then performs subsequent violation identification, which can accurately identify whether the vehicle in the variable lane has violated the law, and there is a basis for judgment, and the judgment result will not be controversial and convincing. In addition, when the vehicle enters the variable lane, the vehicle is associated with the direction of the sign in the variable lane. After entering the variable lane, no matter whether the direction of the sign changes or is damaged (or other reasons cause abnormal display), etc., it will not affect the subsequent judgment of violations, ensuring the effectiveness and accuracy of the illegal driving map collected for the vehicle and reducing the waste map rate.

[0155] Based on the above method embodiment, the present application also provides a vehicle violation identification device, which is applied to a binocular camera, wherein the field of view of the first camera of the binocular camera is the first field of view, and the field of view of the second camera of the binocular camera is the second field of view. The first field of view includes the sign of the variable lane corresponding to the traffic intersection, and the section of the variable lane indicated by the sign, and the second field of view includes the traffic intersection and the lane sections of each lane; Fig.10 As shown, the device comprises:

[0156] The first control module 1010 is used to control the first camera to collect images in response to any vehicle entering the road section of the variable lane within the first field of view, and analyze the collected images to obtain a first feature associated with the vehicle; wherein the first feature associated with any vehicle includes the vehicle feature of the vehicle, and the lane number and sign feature of the variable lane to which the vehicle is traveling; the sign feature is used to represent the driving direction currently indicated in the sign of the variable lane to which the vehicle is traveling;

[0157] The second control module 1020 is used to control the second camera to collect images in response to any vehicle entering the traffic intersection within the second field of view, and analyze the collected images to obtain a second feature associated with the vehicle; wherein the second feature associated with any vehicle includes the vehicle feature of the vehicle, the lane number of the lane to which the vehicle is traveling, and the driving direction feature of the vehicle when passing through the traffic intersection;

[0158] A first determination module 1030 is used to determine, for a target vehicle associated with a second feature, a vehicle that is the same vehicle as the target vehicle from among the vehicles associated with the first feature based on the vehicle feature of the target vehicle and the vehicle features in the first features, and use the first feature of the determined vehicle as a first reference feature;

[0159] The first judgment module 1040 is used to respond to the lane number in the first reference feature being the same as the lane number in the second feature associated with the target vehicle, take the sign feature in the first reference feature as the judgment basis, and based on the driving direction feature in the second feature associated with the target vehicle, judge whether the target vehicle has violated the regulations by not driving according to the guidance.

[0160] The vehicle violation identification device provided in the embodiment of the present application is applied to a binocular camera, and the field of view of the binocular camera includes the first field of view of the first camera and the second field of view of the second camera. The first field of view includes the sign of the variable lane and the corresponding road section, and the second field of view includes the traffic intersection and the lane section of each lane; the field of view of the first camera includes the sign of the variable lane, and the first feature associated with any vehicle obtained by analysis includes the vehicle feature of the vehicle, and the lane number and sign feature of the variable lane driven by the vehicle. The sign feature is used to characterize the current driving direction indicated in the sign of the variable lane driven by the vehicle, that is, to associate the various features of the vehicle when entering the variable lane with the vehicle; the field of view of the second camera includes the traffic intersection and each vehicle The second feature of any vehicle obtained by analysis includes the vehicle feature of the vehicle, the lane number of the lane to which the vehicle is traveling, and the driving direction feature when passing through the traffic intersection, that is, the various features of the vehicle when leaving the lane section and passing through the traffic intersection are associated with the vehicle; for the target vehicle associated with the second feature, through the vehicle feature of the target vehicle and the vehicle features in each first feature, the vehicle identical to the target vehicle can be determined from the vehicles associated with the first feature, and the first reference feature of the target vehicle can be obtained; when the lane number in the first reference feature is the same as the lane number in the second feature associated with the target vehicle, the signboard feature in the first reference feature can be used as the basis for judgment, and based on the driving direction feature associated with the target vehicle, it can be accurately judged whether the target vehicle has violated the rules by not following the guidance. It can be seen that the present application uses two cameras to respectively collect images of their corresponding field of view ranges, and can associate various features when the vehicle enters the variable lane with the vehicle, and associate various features when the vehicle leaves the lane section and passes through the traffic intersection with the vehicle, so that after identifying the same target vehicle, the signboard feature when the target vehicle enters the variable lane can be used as the basis for judgment, and the vehicle in the variable lane can be accurately identified whether there is any violation.

[0161] Optionally, the device further comprises:

[0162] The second judgment module is used for:

[0163] In response to the lane number in the first reference feature being different from the lane number in the second feature associated with the target vehicle, it is determined that the target vehicle has violated the lane change rule by crossing a solid line.

[0164] Optionally, the first feature associated with any vehicle further includes: the driving order of the vehicle in the variable lane to which it is traveling; the second feature associated with any vehicle further includes: the driving order of the vehicle in the lane to which it is traveling;

[0165] The device also includes:

[0166] The third judgment module is used to:

[0167] In response to the lane number in the first reference feature being the same as the lane number in the second feature associated with the target vehicle, it is determined whether the target vehicle has violated the lane change rule by changing lanes along a solid line based on the driving sequence in the first reference feature and the driving sequence in the second feature associated with the target vehicle.

[0168] Optionally, the first feature associated with any vehicle further includes a first time, wherein the first time is a time for performing image acquisition in response to the vehicle entering a lane section of a variable lane within the first field of view;

[0169] The second feature associated with any vehicle further includes a second time, wherein the second time is a time for performing image acquisition in response to the vehicle entering a traffic intersection within the second field of view;

[0170] The device also includes:

[0171] The fourth judgment module is used to:

[0172] In response to the lane number in the first reference feature being the same as the lane number in the second feature associated with the target vehicle, determining whether the target vehicle has committed a parking violation based on a first time in the first reference feature and a second time in the second feature associated with the target vehicle;

[0173] In response to the lane number in the first reference feature being different from the lane number in the second feature associated with the target vehicle, it is determined whether the target vehicle has committed a lane-cutting violation based on a first time in the first reference feature and a second time in the second feature associated with the target vehicle.

[0174] Optionally, the second feature associated with any vehicle further includes: a feature of a signal light at the traffic intersection when the vehicle passes through the traffic intersection;

[0175] The device also includes:

[0176] The fifth judgment module is used to:

[0177] Based on the traffic light feature in the second feature associated with the target vehicle, it is determined whether the target vehicle has a violation of driving without following traffic lights.

[0178] Optionally, the fifth judgment module includes:

[0179] A first judgment submodule is used for judging whether the target vehicle has violated the violation of driving without following the guide based on a first signal light feature in response to judging that the target vehicle has not violated the violation of driving without following the guide; wherein the first signal light feature is a signal light feature about a first signal light in a second feature associated with the target vehicle, and the first signal light is a signal light corresponding to the driving direction indicated by the sign feature in the first reference feature;

[0180] The second judgment submodule is used for judging whether the target vehicle has violated the regulations by not driving according to the guidance based on the second signal light feature in response to judging that the target vehicle has violated the regulations by not driving according to the guidance; wherein the second signal light feature is: the signal light feature for the second signal light in the second feature associated with the target vehicle, and the second signal light is: the signal light corresponding to the driving direction represented by the driving direction feature in the second feature associated with the target vehicle.

[0181] Optionally, the device further comprises:

[0182] a third control module, configured to control the first camera to acquire images in response to any vehicle entering a fixed lane within the first field of view, and to analyze the acquired images to obtain a third feature associated with the vehicle; wherein the third feature associated with any vehicle includes a vehicle feature of the vehicle, and a lane number and a lane driving direction of the fixed lane where the vehicle travels;

[0183] a second determination module, for determining, for a designated vehicle associated with a second feature, a vehicle that is the same vehicle as the designated vehicle from among the vehicles associated with the third feature based on the vehicle feature of the designated vehicle and the vehicle features in the third features, and using the third feature of the determined vehicle as a second reference feature;

[0184] The sixth judgment module is used to judge whether the designated vehicle has violated the regulations based on the lane number in the second reference feature and the lane number and driving direction features in the second feature of the designated vehicle, taking the lane driving direction in the second reference feature as a judgment basis.

[0185] Optionally, the first determination module is specifically configured to:

[0186] If the direction represented by the driving direction feature in the second feature associated with the target vehicle is different from the driving direction indicated by the sign feature in the first reference feature, it is determined that the target vehicle has violated the regulations by not driving according to the guidance.

[0187] The embodiment of the present application further provides a binocular camera, the binocular camera comprising a first camera, a second camera and a main controller; the field of view of the first camera of the binocular camera is a first field of view, the field of view of the second camera of the binocular camera is a second field of view, the first field of view includes a sign of a variable lane corresponding to a traffic intersection, and a road section of the variable lane indicated by the sign, and the second field of view includes the traffic intersection and the lane sections of each lane;

[0188] The first camera is used to collect images of the first field of view under the control of the main controller;

[0189] The second camera is used to collect images of the second field of view under the control of the main controller;

[0190] The main controller is used to control the first camera to collect images of the first field of view in response to any vehicle entering the road section of the variable lane within the first field of view, and analyze the collected images to obtain the first feature associated with the vehicle; in response to any vehicle entering the traffic intersection within the second field of view, control the second camera to collect images of the second field of view, and analyze the collected images to obtain the second feature associated with the vehicle; for a target vehicle associated with the second feature, based on the second vehicle feature of the target vehicle and each first vehicle feature, determine the vehicle that is the same as the target vehicle from each vehicle associated with the first feature, and use the first feature of the determined vehicle as the first reference feature; in response to the first lane number in the first reference feature being the same as the second lane number in the second feature associated with the target vehicle, use the driving direction indicated by the signboard feature in the first reference feature as the judgment basis, and based on the driving direction feature in the second feature associated with the target vehicle, judge whether the target vehicle has violated the rules of not driving according to the guidance;

[0191] Among them, the first feature associated with any vehicle includes: the first vehicle feature of the vehicle, the first lane number of the variable lane in which the vehicle is traveling, and the signboard feature of the variable lane; the signboard feature is used to indicate the driving direction of the variable lane in which the vehicle is traveling; the second feature associated with any vehicle includes: the second vehicle feature of the vehicle, the second lane number of the lane in which the vehicle is traveling, and the driving direction feature of the vehicle.

[0192] The binocular camera provided in the embodiment of the present application has a field of view including a first field of view of a first camera and a second field of view of a second camera, the first field of view includes the sign of the variable lane and the corresponding road section, and the second field of view includes the traffic intersection and the road section of each lane; the field of view of the first camera includes the sign of the variable lane, and the first feature associated with any vehicle obtained by analysis includes the vehicle feature of the vehicle, and the lane number and sign feature of the variable lane traveled by the vehicle, and the sign feature is used to characterize the current driving direction indicated in the sign of the variable lane traveled by the vehicle, that is, to associate various features when the vehicle enters the variable lane with the vehicle; the field of view of the second camera includes the traffic intersection and the road section of each lane, The second feature of any vehicle obtained by analysis includes the vehicle feature of the vehicle, the lane number of the lane to which the vehicle is traveling, and the driving direction feature when passing through the traffic intersection, that is, the various features of the vehicle when leaving the road section and passing through the traffic intersection are associated with the vehicle; for the target vehicle associated with the second feature, through the vehicle feature of the target vehicle and the vehicle features in each first feature, the vehicle identical to the target vehicle can be determined from the vehicles associated with the first feature, and the first reference feature of the target vehicle can be obtained; when the lane number in the first reference feature is the same as the lane number in the second feature associated with the target vehicle, the signboard feature in the first reference feature can be used as a basis for judgment, and based on the driving direction feature associated with the target vehicle, it can be accurately judged whether the target vehicle has violated the rules by not driving according to the guide. It can be seen that the present application uses two cameras to respectively collect images of their corresponding field of view ranges, and can associate various features when the vehicle enters the variable lane with the vehicle, and associate various features when the vehicle leaves the road section and passes through the traffic intersection with the vehicle, so that after identifying the same target vehicle, the signboard feature when the target vehicle enters the variable lane can be used as a basis for judgment, and the vehicle in the variable lane can be accurately identified whether there is any violation.

[0193] The present application also provides an electronic device, such as Fig.11 As shown, including:

[0194] Memory 1101, used for storing computer programs;

[0195] The processor 1102 is used to implement any of the above-mentioned vehicle violation identification methods when executing the program stored in the memory 1101.

[0196] In addition, the electronic device may further include a communication bus and / or a communication interface, and the processor 1102, the communication interface, and the memory 1101 communicate with each other via the communication bus.

[0197] The communication bus mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0198] The communication interface is used for communication between the above electronic device and other devices.

[0199] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0200] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0201] In another embodiment provided in the present application, a computer-readable storage medium is also provided, in which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned vehicle violation identification methods are implemented.

[0202] In another embodiment provided in the present application, a computer program product including instructions is also provided, which, when executed on a computer, enables the computer to execute any of the vehicle violation identification methods in the above embodiments.

[0203] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a solid-state hard disk (SSD), etc.

[0204] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0205] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0206] The above description is only a preferred embodiment of the present application and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application are included in the protection scope of the present application.

Claims

1. A method for identifying vehicle violation behavior, characterized in that: Applied to a binocular camera, the field of view of the first camera of the binocular camera is the first field of view, the field of view of the second camera of the binocular camera is the second field of view, the first field of view includes a sign of a variable lane corresponding to a traffic intersection, and a section of the variable lane indicated by the sign, and the direction indicated by the sign includes going straight or turning; The second field of view includes the traffic intersection and the lane sections of each lane; The method comprises: In response to any vehicle entering a road section of a variable lane within the first field of view, the first camera is controlled to collect images, and the collected images are analyzed to obtain a first feature associated with the vehicle; wherein the first feature associated with any vehicle includes a vehicle feature of the vehicle, and a lane number and a sign feature of the variable lane to which the vehicle is traveling; the sign feature is used to represent the driving direction currently indicated in the sign of the variable lane to which the vehicle is traveling; In response to any vehicle entering the second field of view, controlling the second camera to collect images, performing vehicle feature analysis on the collected images, matching the vehicle features obtained by the analysis with the vehicle features in each first feature, and obtaining an analysis result of the vehicle; In the case where the vehicle has an analysis result indicating successful matching, in response to the vehicle entering a traffic intersection within the second field of view, controlling the second camera to acquire an image, and analyzing the acquired image to obtain a second feature associated with the vehicle; wherein the second feature associated with any vehicle includes a vehicle feature of the vehicle, a lane number of a lane to which the vehicle is traveling, a driving direction feature of the vehicle when passing through the traffic intersection, and a signal light feature of the traffic intersection; For a target vehicle associated with a second feature, based on the vehicle feature of the target vehicle and the vehicle features in each of the first features, a vehicle that is the same as the target vehicle is determined from each of the vehicles associated with the first feature, and the first feature of the determined vehicle is used as a first reference feature; In response to the lane number in the first reference feature being the same as the lane number in the second feature associated with the target vehicle, judging whether the target vehicle has violated the driving rules by not following the driving directions based on the sign feature in the first reference feature and the driving direction feature in the second feature associated with the target vehicle; Based on the traffic light feature in the second feature associated with the target vehicle, determine whether the target vehicle has committed a violation of not driving according to the traffic light; wherein, determining whether the target vehicle has committed a violation of not driving according to the traffic light includes: based on the first traffic light feature, determining whether the target vehicle has committed a violation of not driving according to the traffic light; wherein, the first traffic light feature is a traffic light feature regarding a first traffic light in the second feature associated with the target vehicle, and the first traffic light is a traffic light corresponding to the driving direction indicated by the sign feature in the first reference feature.

2. The method according to claim 1, characterized in that The method further comprises: In response to the lane number in the first reference feature being different from the lane number in the second feature associated with the target vehicle, it is determined that the target vehicle has violated the lane change rule by crossing a solid line.

3. The method according to claim 1, characterized in that The first feature associated with any vehicle also includes: the driving order of the vehicle in the variable lane it is traveling to; the second feature associated with any vehicle also includes: the driving order of the vehicle in the lane it is traveling to; The method further comprises: In response to the lane number in the first reference feature being the same as the lane number in the second feature associated with the target vehicle, it is determined whether the target vehicle has violated the lane change rule by changing lanes along a solid line based on the driving sequence in the first reference feature and the driving sequence in the second feature associated with the target vehicle.

4. The method according to claim 1, characterized in that: The first feature associated with any vehicle further includes a first time, wherein the first time is a time for performing image acquisition in response to the vehicle entering a lane section of a variable lane within the first field of view; The second feature associated with any vehicle further includes a second time, wherein the second time is a time for performing image acquisition in response to the vehicle entering a traffic intersection within the second field of view; The method further comprises: In response to the lane number in the first reference feature being the same as the lane number in the second feature associated with the target vehicle, determining whether the target vehicle has committed a parking violation based on a first time in the first reference feature and a second time in the second feature associated with the target vehicle; In response to the lane number in the first reference feature being different from the lane number in the second feature associated with the target vehicle, it is determined whether the target vehicle has committed a lane-cutting violation based on a first time in the first reference feature and a second time in the second feature associated with the target vehicle.

5. The method according to any one of claims 1 to 4, characterized in that: The determining, based on the signal light feature in the second feature associated with the target vehicle, whether the target vehicle has violated the traffic rules by not following the signal light, further includes: In response to determining that the target vehicle has violated the regulations by not driving according to the guidance, a determination is made based on the second signal light feature whether the target vehicle has violated the regulations by not driving according to the signal light; wherein the second signal light feature is: the signal light feature for the second signal light in the second feature associated with the target vehicle, and the second signal light is: the signal light corresponding to the driving direction represented by the driving direction feature in the second feature associated with the target vehicle.

6. The method according to claim 1, characterized in that The method further comprises: In response to any vehicle entering a fixed lane within the first field of view, controlling the first camera to collect images, and analyzing the collected images to obtain a third feature associated with the vehicle; wherein the third feature associated with any vehicle includes a vehicle feature of the vehicle, and a lane number and lane driving direction of the fixed lane where the vehicle travels; For a designated vehicle associated with the second feature, based on the vehicle feature of the designated vehicle and the vehicle features in each third feature, a vehicle that is the same vehicle as the designated vehicle is determined from each vehicle associated with the third feature, and the third feature of the determined vehicle is used as the second reference feature; Taking the lane driving direction in the second reference feature as a judgment basis, based on the lane number in the second reference feature, and the lane number and driving direction features in the second feature of the designated vehicle, it is judged whether the designated vehicle has violated the regulations.

7. The method according to claim 1, characterized in that The step of judging whether the target vehicle has violated the rules of not driving according to the guidance based on the sign feature in the first reference feature and the driving direction feature in the second feature associated with the target vehicle includes: If the direction represented by the driving direction feature in the second feature associated with the target vehicle is different from the driving direction indicated by the sign feature in the first reference feature, it is determined that the target vehicle has violated the regulations by not driving according to the guidance.

8. A vehicle violation identification device, characterized in that: Applied to a binocular camera, the field of view of the first camera of the binocular camera is the first field of view, the field of view of the second camera of the binocular camera is the second field of view, the first field of view includes a sign of a variable lane corresponding to a traffic intersection, and a section of the variable lane indicated by the sign, and the direction indicated by the sign includes going straight or turning; The second field of view includes the traffic intersection and the lane sections of each lane; The device comprises: a first control module, configured to control the first camera to acquire images in response to any vehicle entering a road section of a variable lane within the first field of view, and to analyze the acquired images to obtain a first feature associated with the vehicle; wherein the first feature associated with any vehicle includes a vehicle feature of the vehicle, and a lane number and a sign feature of the variable lane to which the vehicle is traveling; the sign feature is used to represent a driving direction currently indicated by the sign of the variable lane to which the vehicle is traveling; A second control module is used to control the second camera to collect images in response to any vehicle entering the second field of view, perform vehicle feature analysis on the collected images, match the vehicle features obtained by analysis with the vehicle features in each first feature, and obtain an analysis result of the vehicle; in response to the vehicle entering the traffic intersection within the second field of view, control the second camera to collect images and analyze the collected images to obtain a second feature associated with the vehicle, wherein the second feature associated with any vehicle includes the vehicle feature of the vehicle, the lane number of the lane to which the vehicle is traveling, and the driving direction feature of the vehicle when passing through the traffic intersection and the signal light feature of the traffic intersection; A first determination module is used to determine, for a target vehicle associated with a second feature, a vehicle that is the same vehicle as the target vehicle from among the vehicles associated with the first feature based on the vehicle feature of the target vehicle and the vehicle features in the first features, and use the first feature of the determined vehicle as a first reference feature; a first judgment module, configured to, in response to the lane number in the first reference feature being the same as the lane number in the second feature associated with the target vehicle, use the sign feature in the first reference feature as a judgment basis, and based on the driving direction feature in the second feature associated with the target vehicle, judge whether the target vehicle has violated the driving rules by not following the guidance; a fifth judgment module, for judging whether the target vehicle has committed a violation of not driving according to the signal light based on the signal light feature in the second feature associated with the target vehicle; wherein, judging whether the target vehicle has committed a violation of not driving according to the signal light comprises: judging whether the target vehicle has committed a violation of not driving according to the signal light based on the first signal light feature; wherein, the first signal light feature is a signal light feature regarding a first signal light in the second feature associated with the target vehicle, and the first signal light is a signal light corresponding to the driving direction indicated by the sign feature in the first reference feature.

9. A binocular camera, characterized in that: The binocular camera includes a first camera, a second camera and a main controller; the field of view of the first camera of the binocular camera is a first field of view, and the field of view of the second camera of the binocular camera is a second field of view. The first field of view includes a sign of a variable lane corresponding to a traffic intersection, and a section of the variable lane indicated by the sign, and the direction indicated by the sign includes going straight or turning; the second field of view includes the traffic intersection and the lane sections of each lane; The first camera is used to collect images of the first field of view under the control of the main controller; The second camera is used to collect images of the second field of view under the control of the main controller; The main controller is configured to control the first camera to collect images of the first field of view in response to any vehicle entering the lane section of the variable lane within the first field of view, and analyze the collected images to obtain a first feature associated with the vehicle; In response to any vehicle entering the second field of view, the second camera is controlled to collect images, vehicle feature analysis is performed on the collected images, and the vehicle features obtained by the analysis are matched with the vehicle features in each first feature to obtain an analysis result of the vehicle; in the case where the vehicle has an analysis result indicating a successful match, in response to the vehicle entering a traffic intersection within the second field of view, the second camera is controlled to collect images within the second field of view, and the collected images are analyzed to obtain a second feature associated with the vehicle; For a target vehicle associated with a second feature, based on the second vehicle feature of the target vehicle and each first vehicle feature, a vehicle that is the same as the target vehicle is determined from each vehicle associated with the first feature, and the first feature of the determined vehicle is used as a first reference feature; In response to the first lane number in the first reference feature being the same as the second lane number in the second feature associated with the target vehicle, taking the driving direction indicated by the sign feature in the first reference feature as a basis for determination, and based on the driving direction feature in the second feature associated with the target vehicle, determining whether the target vehicle has violated the driving rules by not following the guidance; The main controller is further used to determine whether the target vehicle has a violation of driving without following the signal light based on the signal light feature in the second feature associated with the target vehicle; wherein the determination of whether the target vehicle has a violation of driving without following the signal light comprises: determining whether the target vehicle has a violation of driving without following the signal light based on the first signal light feature; wherein the first signal light feature is a signal light feature about a first signal light in the second feature associated with the target vehicle, and the first signal light is a signal light corresponding to the driving direction indicated by the sign feature in the first reference feature; Among them, the first feature associated with any vehicle includes: the first vehicle feature of the vehicle, the first lane number of the variable lane in which the vehicle is traveling, and the signboard feature of the variable lane in which the vehicle is traveling; the signboard feature is used to indicate the driving direction of the variable lane in which the vehicle is traveling; the second feature associated with any vehicle includes: the second vehicle feature of the vehicle, the second lane number of the lane in which the vehicle is traveling, the driving direction feature of the vehicle and the traffic light feature of the traffic intersection.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Citation Information

Patent Citations

  • Traffic video behavior analyzing and alarming server

    CN102081844A

  • Vehicle video monitoring method, server and system

    CN107305627A

  • Traffic illegal behavior identification method and device

    CN108847032A

  • AI (artificial intelligence) high-altitude point panorama-based intelligent traffic violation monitoring method and system

    CN109859487A

  • Vehicle violation video processing method and device, computer equipment and storage medium

    CN110533925A