Method, system, device, equipment and medium for collecting evidence of vehicle lane-changing behavior on a solid line

Abstract

The embodiment of the present application provides a method, system, device, equipment and medium for collecting evidence of a vehicle's solid line lane change behavior, which relates to the field of traffic management technology. The method is applied to a data processing end, including: receiving vehicle snapshot information reported by a first camera and a second camera; obtaining a first virtual queue and a second virtual queue; according to the coordinates of the position frame in the snapshot information and the coordinates of each position frame in the virtual queue corresponding to the camera that reported the snapshot information, associating the first category ID of the position frame in the snapshot information with the second category ID belonging to the same vehicle; determining the second category ID of the position frame of the vehicle in the illegal image as the target ID, searching for the first category ID associated with the target ID, and generating the evidence collection result of the vehicle based on the illegal image and the vehicle image in the snapshot information to which the found first category ID belongs. Through this solution, it is possible to collect evidence of vehicles that have undergone solid line lane change behavior.

Description

Technical Field

[0001] The present application relates to the field of traffic management technology, and in particular to a method, system, device, equipment and medium for collecting evidence of a vehicle's lane-changing behavior on a solid line. Background Art

[0002] Vehicles must meet driving regulations during driving, for example, vehicles cannot change lanes on solid-line roads. In order to reduce vehicle violations, it is usually necessary to identify vehicle violations.

[0003] To deal with lane changing on a solid line, currently, electric police cameras are set up near the stop line of the solid line road at the traffic intersection. The electric police cameras are used to identify and collect evidence of the vehicle's lane changing on a solid line.

[0004] However, due to the blind spot below the electric police camera, the electric police camera cannot collect evidence of vehicles that change lanes on a solid line within the blind spot. Summary of the invention

[0005] The purpose of the embodiments of the present application is to provide a method, system, device, equipment and medium for collecting evidence of a vehicle changing lanes on a solid line, so as to collect evidence of a vehicle changing lanes on a solid line. The specific technical solution is as follows:

[0006] In a first aspect, an embodiment of the present application provides a method for collecting evidence of a solid-line lane change behavior of a vehicle, which is applied to a data processing end, wherein the data processing end communicates with a first camera, a fisheye camera, and a second camera, wherein the field of view of the first camera is the direction of the oncoming vehicle of each solid-line road, the field of view of the fisheye camera is the top-down direction of each solid-line road, the field of view of the second camera is the outgoing direction of each solid-line road, and the fisheye camera has overlapping acquisition areas with the first camera and the second camera respectively; the method comprises:

[0007] Receive each captured information reported by each camera of the first camera and the second camera; wherein each captured information reported by each camera includes: a vehicle image captured by the camera when the camera recognizes that a vehicle enters an overlapping acquisition area with the fisheye camera, and a first type ID and coordinates of a position frame of the vehicle in the vehicle image;

[0008] Acquire a first virtual queue and a second virtual queue corresponding to each frame image; wherein each frame image is collected by the fisheye camera in the process of tracking vehicles entering the field of view, and each frame image in each frame image corresponds to a vehicle queue of the second type ID and coordinates of the position box of each vehicle in the image; the first virtual queue corresponding to each frame image is a vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into coordinates in the image coordinate system of the first camera; the second virtual queue corresponding to each frame image is a vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into coordinates in the image coordinate system of the second camera;

[0009] For each received snapshot information, according to the coordinates of the location frame in the snapshot information and the coordinates of each location frame in the virtual queue corresponding to the camera reporting the snapshot information, the first category ID of the location frame in the snapshot information is associated with the second category ID belonging to the same vehicle; wherein the virtual queue corresponding to the first camera is the first virtual queue, and the virtual queue corresponding to the second camera is the second virtual queue;

[0010] In response to receiving an illegal image of any vehicle reported by the fisheye camera, a second-category ID of a position box of the vehicle in the illegal image is determined as a target ID, a first-category ID associated with the target ID is searched, and based on the illegal image and the vehicle image in the captured information to which the found first-category ID belongs, a forensic result of the vehicle's lane changing behavior on a solid line is generated; wherein the illegal image is an image captured by the fisheye camera when it is identified that any vehicle has committed an illegal lane changing behavior.

[0011] In a second aspect, an embodiment of the present application provides a system for collecting evidence of a solid-line lane-changing behavior of a vehicle, comprising: a data processing end, a first camera, a second camera, and a fisheye camera; the data processing end communicates with the first camera, the fisheye camera, and the second camera, the field of view of the first camera is the direction of the oncoming vehicle of each solid-line road, the field of view of the fisheye camera is the top-down direction of each solid-line road, the field of view of the second camera is the outgoing direction of each solid-line road, and the fisheye camera has overlapping acquisition areas with the first camera and the second camera respectively;

[0012] Each of the first camera and the second camera is used to capture the vehicle when recognizing that a vehicle enters the overlapping collection area with the fisheye camera, and report the captured information obtained by capturing the vehicle to the data processing end; wherein the captured information includes a captured vehicle image, a first type ID and coordinates of a position frame of the vehicle in the vehicle image;

[0013] The fisheye camera is used to track vehicles entering the field of view, and perform image analysis and processing on each frame of images collected during the tracking process to obtain a vehicle queue of the second type ID and coordinates of a position frame of each vehicle in the image corresponding to each frame of images;

[0014] The data processing end is used to receive each snapshot information reported by each of the first camera and the second camera, and obtain the first virtual queue and the second virtual queue corresponding to each frame image; and, for each snapshot information received, according to the coordinates of the position frame in the snapshot information and the coordinates of each position frame in the virtual queue corresponding to the camera reporting the snapshot information, associate the first category ID of the position frame in the snapshot information with the second category ID belonging to the same vehicle; wherein the virtual queue corresponding to the first camera is the first virtual queue, and the virtual queue corresponding to the second camera is the second virtual queue; the first virtual queue corresponding to each frame image is the vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into the coordinates in the image coordinate system of the first camera; the second virtual queue corresponding to each frame image is the vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into the coordinates in the image coordinate system of the second camera;

[0015] The fisheye camera is further used to collect images when identifying any vehicle that has made an illegal lane change, obtain an illegal image of the vehicle, and report the illegal image to the data processing end;

[0016] The data processing end is also used to receive the violation image, determine the second category ID of the position box of the vehicle in the violation image as the target ID, search for the first category ID associated with the target ID, and generate the evidence results of the vehicle's lane changing behavior on the solid line based on the violation image and the vehicle image in the captured information to which the found first category ID belongs.

[0017] In a third aspect, an embodiment of the present application provides a device for collecting evidence of a solid-line lane change behavior of a vehicle, which is applied to a data processing end, wherein the data processing end communicates with a first camera, a fisheye camera, and a second camera, wherein the field of view of the first camera is the direction of the oncoming vehicle of each solid-line road, the field of view of the fisheye camera is the top-down direction of each solid-line road, the field of view of the second camera is the outgoing direction of each solid-line road, and the fisheye camera has overlapping acquisition areas with the first camera and the second camera respectively; the device comprises:

[0018] A first receiving module is used to receive each captured information reported by each camera of the first camera and the second camera; wherein each captured information reported by each camera includes: a vehicle image captured by the camera when the camera recognizes that a vehicle enters the overlapping acquisition area with the fisheye camera, and a first type ID and coordinates of a position frame of the vehicle in the vehicle image;

[0019] an acquisition module, for acquiring a first virtual queue and a second virtual queue corresponding to each frame image; wherein each frame image is collected by the fisheye camera in the process of tracking vehicles entering the field of view, and each frame image in each frame image corresponds to a vehicle queue of the second type ID and coordinates of a position frame containing each vehicle in the image; the first virtual queue corresponding to each frame image is a vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into coordinates in the image coordinate system of the first camera; the second virtual queue corresponding to each frame image is a vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into coordinates in the image coordinate system of the second camera;

[0020] an associating module, for associating, for each received snapshot information, a first category ID of the location frame in the snapshot information with a second category ID belonging to the same vehicle according to the coordinates of the location frame in the snapshot information and the coordinates of each location frame in the virtual queue corresponding to the camera reporting the snapshot information; wherein the virtual queue corresponding to the first camera is the first virtual queue, and the virtual queue corresponding to the second camera is the second virtual queue;

[0021] A generation module is used to, in response to receiving an illegal image of any vehicle reported by the fisheye camera, determine the second category ID of the position box of the vehicle in the illegal image as the target ID, search for the first category ID associated with the target ID, and generate a forensic result of the vehicle's lane changing behavior on a solid line based on the illegal image and the vehicle image in the captured information to which the found first category ID belongs; wherein the illegal image is an image captured by the fisheye camera when it is identified that any vehicle has made an illegal lane changing behavior.

[0022] In a fourth aspect, an embodiment of the present application provides a data processing device, the data processing device communicates with a first camera, a fisheye camera, and a second camera, the field of view of the first camera is the direction of the vehicle coming from each solid-line road, the field of view of the fisheye camera is the top-down direction of each solid-line road, the field of view of the second camera is the direction of the vehicle going from each solid-line road, and the fisheye camera has overlapping acquisition areas with the first camera and the second camera respectively; the data processing device includes:

[0023] Memory, used to store computer programs;

[0024] The processor is used to implement the method for collecting evidence of the vehicle's solid line lane change behavior when executing the program stored in the memory.

[0025] In a fifth aspect, an embodiment of the present application provides an all-in-one device, comprising a first camera, a fisheye camera, a second camera, and a data processing end; the data processing end communicates with the first camera, the fisheye camera, and the second camera, the field of view of the first camera is the direction of the vehicle coming from each solid-line road, the field of view of the fisheye camera is the direction of looking down at each solid-line road, the field of view of the second camera is the direction of the vehicle going out of each solid-line road, and the fisheye camera has overlapping acquisition areas with the first camera and the second camera respectively;

[0026] Each of the first camera and the second camera is used to capture the vehicle when recognizing that a vehicle enters the overlapping collection area with the fisheye camera, and report the captured information obtained by capturing the vehicle to the data processing end; wherein the captured information includes a captured vehicle image, a first type ID and coordinates of a position frame of the vehicle in the vehicle image;

[0027] The fisheye camera is used to track vehicles entering the field of view, and perform image analysis and processing on each frame of images collected during the tracking process to obtain a vehicle queue of the second type ID and coordinates of a position frame of each vehicle in the image corresponding to each frame of images;

[0028] The data processing end is used to receive each snapshot information reported by each of the first camera and the second camera, and obtain the first virtual queue and the second virtual queue corresponding to each frame image; and, for each snapshot information received, according to the coordinates of the position frame in the snapshot information and the coordinates of each position frame in the virtual queue corresponding to the camera reporting the snapshot information, associate the first category ID of the position frame in the snapshot information with the second category ID belonging to the same vehicle; wherein the virtual queue corresponding to the first camera is the first virtual queue, and the virtual queue corresponding to the second camera is the second virtual queue; the first virtual queue corresponding to each frame image is the vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into the coordinates in the image coordinate system of the first camera; the second virtual queue corresponding to each frame image is the vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into the coordinates in the image coordinate system of the second camera;

[0029] The fisheye camera is further used to collect images when identifying any vehicle that has made an illegal lane change, obtain an illegal image of the vehicle, and report the illegal image to the data processing end;

[0030] The data processing end is also used to receive the violation image, determine the second category ID of the position box of the vehicle in the violation image as the target ID, search for the first category ID associated with the target ID, and generate the evidence results of the vehicle's lane changing behavior on the solid line based on the violation image and the vehicle image in the captured information to which the found first category ID belongs.

[0031] In a sixth aspect, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the method for collecting evidence of a vehicle's solid line lane changing behavior is implemented.

[0032] 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-mentioned methods for collecting evidence of a vehicle changing lanes along a solid line.

[0033] Beneficial effects of the embodiments of the present application:

[0034] The method for collecting evidence of a vehicle's solid-line lane-changing behavior provided in an embodiment of the present application is applied to a data processing end, and the data processing end communicates with a first camera, a fisheye camera, and a second camera. The field of view of the first camera is the oncoming direction of each solid-line road, the field of view of the fisheye camera is the overhead direction of each solid-line road, and the field of view of the second camera is the outgoing direction of each solid-line road. The fisheye camera has overlapping acquisition areas with the first camera and the second camera, respectively, so that when collecting evidence of a vehicle that has changed lanes through a solid line through the first camera, the fisheye camera, and the second camera, there is no blind spot for evidence collection. Specifically, the data processing end can receive each snapshot information reported by each of the first camera and the second camera, and obtain the first virtual queue and the second virtual queue corresponding to each frame image; for each snapshot information received, according to the coordinates of the position frame in the snapshot information and the coordinates of each position frame in the virtual queue corresponding to the camera reporting the snapshot information, the first category ID of the position frame in the snapshot information is associated with the second category ID belonging to the same vehicle, the virtual queue corresponding to the first camera is the first virtual queue, and the virtual queue corresponding to the second camera is the second virtual queue, which is equivalent to, for each snapshot information received and reported by the first camera and the second camera, through the coordinates of the position frame, The first category ID of the captured information is associated with the second category ID of the same vehicle in the corresponding virtual queue reported by the fisheye camera, that is, the position of the same vehicle in the field of view of the oncoming direction of each solid road is associated with the position in the field of view of the overlooking direction, and the position in the field of view of the outgoing direction of each solid road is associated with the position in the field of view of the overlooking direction; thus, in response to receiving the illegal image of any vehicle reported by the fisheye camera, the data processing end can determine the second category ID of the position frame of the vehicle in the illegal image as the target ID, and search for the first category ID associated with the target ID, and generate the evidence result of the vehicle's solid line lane change behavior based on the vehicle image and the illegal image in the captured information to which the first category ID belongs. It can be seen that in this application, the field of view of the fisheye camera is the overlooking direction of each solid road, and the illegal image of any vehicle can be collected, and combined with the vehicle image in the captured information to which the first category ID of the target ID belongs, that is, combined with the captured image of the vehicle in the field of view of the oncoming direction and the captured image in the field of view of the outgoing direction, a complete evidence result about the solid line lane change behavior is generated for the vehicle. This solution can be used to collect evidence from vehicles that change lanes on a solid line.

[0035] 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

[0036] 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.

[0037] Figure 1 A schematic diagram of a process for collecting evidence of a vehicle's solid lane change behavior provided in an embodiment of the present application;

[0038] Figure 2 A schematic diagram of the field of view of a first camera, a fisheye camera, and a second camera provided in an embodiment of the present application;

[0039] Figure 3 A schematic diagram of the field of view of a first camera, a fisheye camera, and a second camera provided in an embodiment of the present application;

[0040] Figure 4 A schematic diagram of a system for collecting evidence of a vehicle's solid lane change behavior provided in an embodiment of the present application;

[0041] Figure 5 A schematic diagram of the interaction process of a system for collecting evidence of a vehicle's solid lane change behavior provided in an embodiment of the present application;

[0042] Figure 6 A schematic diagram of a process of mapping a vehicle queue of each frame image of a fisheye provided in an embodiment of the present application to a checkpoint and reporting to a data processing terminal;

[0043] Figure 7 A schematic diagram of a process for associating the same vehicle by ID provided in an embodiment of the present application;

[0044] Figure 8 A schematic diagram of the process of generating evidence collection results by image synthesis provided in an embodiment of the present application;

[0045] Fig. 9 A schematic diagram of an illegal lane change capture image captured by a fisheye camera provided in an embodiment of the present application;

[0046] Fig.10 A schematic diagram of a composite image obtained by synthesizing images provided in an embodiment of the present application;

[0047] Fig.11 A schematic diagram of the structure of a device for collecting evidence of a vehicle's solid lane change behavior provided in an embodiment of the present application;

[0048] Fig.12 A block diagram of a data processing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] 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.

[0050] When collecting evidence of a vehicle that changes lanes on a solid line, there is a scheme in the related art for collecting evidence by using a checkpoint camera, an electric police camera, and a fisheye camera to capture illegal lane changes in a blind spot. Specifically, this scheme can capture images of the vehicle's illegal lane change by using a checkpoint camera and an electric police camera, and the obtained evidence results include the captured images of the checkpoint and the captured images of the electric police. The fisheye camera can only record but cannot capture images, and the obtained evidence results still do not include the blind spot below the electric police camera. , a captured image of the vehicle's actual solid line lane change behavior; and, when matching the captured images captured by the checkpoint camera and the electric police camera about the same vehicle, the license plate number in the captured image of the checkpoint and the license plate number in the captured image of the electric police are used to determine whether they are the same vehicle. In some scenarios, such as: the vehicle has no license plate, or the license plate is blocked by the front and rear vehicles, making the license plate unrecognizable, or a large vehicle cannot reveal its license plate (such as the rear license plate), this solution cannot collect evidence for vehicles that have changed lanes on a solid line.

[0051] Based on this, the present application provides a method, system, device, equipment and medium for collecting evidence of a vehicle's solid line lane changing behavior, so as to collect evidence of a vehicle that has changed lanes on a solid line.

[0052] The following is an introduction to the evidence collection method for vehicle solid line lane changing behavior provided by this application.

[0053] Among them, the evidence collection method for the solid line lane change behavior of the vehicle provided in the embodiment of the present application can be applied to the data processing end, and the data processing end can be a mobile phone or a computer, etc., and the present application does not limit the specific form of the data processing end. In addition, the data processing end communicates with the first camera, the fisheye camera and the second camera. The field of view of the first camera is the direction of the oncoming vehicle of each solid line road, the field of view of the fisheye camera is the top-down direction of each solid line road, and the field of view of the second camera is the outgoing direction of each solid line road. The fisheye camera has overlapping acquisition areas with the first camera and the second camera respectively, so that when the first camera, the fisheye camera and the second camera are used to collect evidence of the vehicle that has changed lanes on the solid line, there is no blind spot for evidence collection. It should be noted that the field of view of the first camera may include the starting position of the solid-line road, and the field of view of the second camera may include the ending position of the solid-line road. For example, the field of view of the second camera may include the stop line of the solid-line road at the traffic intersection, etc., and the field of view of the fisheye camera is the overhead direction of each solid-line road, where each road can be a solid-line road, and this application does not limit this.

[0054] It should be emphasized that the data processing end can be integrated into a device together with the first camera, the fisheye camera and the second camera, that is, the data processing end and the first camera, the fisheye camera and the second camera are an all-in-one device, and the data processing end is a module in the all-in-one device, which can communicate with each camera; or, the first camera, the fisheye camera and the second camera are integrated into an integrated device at the same time, the data processing end is a data processing device, and communicates with each camera in the integrated device; or, the data processing end is a separate data processing device, and the first camera, the fisheye camera and the second camera are also separate devices, that is, the data processing end, the first camera, the fisheye camera and the second camera are all separate devices, and the data processing device communicates with the devices of the three cameras respectively; under the condition that the data processing end communicates with the first camera, the fisheye camera and the second camera, the present application does not limit the integration status of each device.

[0055] In addition, the method for collecting evidence of a vehicle's solid line lane changing behavior provided in the embodiments of the present application can be applied to any scenario where there is a need to collect evidence of a vehicle that has performed a solid line lane changing behavior, for example: collecting evidence of the vehicle's solid line lane changing behavior for any pair of solid line road sections between the starting position and the ending position of the solid line road, such as collecting evidence for the solid line road section between the starting position of the solid line road and the ending position of the solid line road at a traffic intersection; the present application does not limit the specific application scenarios of the method for collecting evidence of a vehicle's solid line lane changing behavior.

[0056] In the embodiment of the present application, the first camera and the second camera can be any camera that can perform vehicle identification and image acquisition, and the first camera and the second camera can be the same or different. For example, the first camera and the second camera are both bayonet cameras, or the first camera and the second camera are both electric police cameras. In order to meet the actual scenario, the first camera can be a bayonet camera and the second camera can be an electric police camera.

[0057] Specifically, an embodiment of the present application provides a method for collecting evidence of a solid-line lane change behavior of a vehicle, which is applied to a data processing end, wherein the data processing end communicates with a first camera, a fisheye camera, and a second camera, wherein the field of view of the first camera is the direction of the oncoming vehicle of each solid-line road, the field of view of the fisheye camera is the top-down direction of each solid-line road, the field of view of the second camera is the outgoing direction of each solid-line road, and the fisheye camera has overlapping acquisition areas with the first camera and the second camera respectively; the method comprises:

[0058] Receive each captured information reported by each camera of the first camera and the second camera; wherein each captured information reported by each camera includes: a vehicle image captured by the camera when the camera recognizes that a vehicle enters an overlapping acquisition area with the fisheye camera, and a first type ID and coordinates of a position frame of the vehicle in the vehicle image;

[0059] Acquire a first virtual queue and a second virtual queue corresponding to each frame image; wherein each frame image is collected by the fisheye camera in the process of tracking vehicles entering the field of view, and each frame image in each frame image corresponds to a vehicle queue of the second type ID and coordinates of the position box of each vehicle in the image; the first virtual queue corresponding to each frame image is a vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into coordinates in the image coordinate system of the first camera; the second virtual queue corresponding to each frame image is a vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into coordinates in the image coordinate system of the second camera;

[0060] For each received snapshot information, according to the coordinates of the location frame in the snapshot information and the coordinates of each location frame in the virtual queue corresponding to the camera reporting the snapshot information, the first category ID of the location frame in the snapshot information is associated with the second category ID belonging to the same vehicle; wherein the virtual queue corresponding to the first camera is the first virtual queue, and the virtual queue corresponding to the second camera is the second virtual queue;

[0061] In response to receiving an illegal image of any vehicle reported by the fisheye camera, a second-category ID of a position box of the vehicle in the illegal image is determined as a target ID, a first-category ID associated with the target ID is searched, and based on the illegal image and the vehicle image in the captured information to which the found first-category ID belongs, a forensic result of the vehicle's lane changing behavior on a solid line is generated; wherein the illegal image is an image captured by the fisheye camera when it is identified that any vehicle has committed an illegal lane changing behavior.

[0062] The method for collecting evidence of a vehicle's solid-line lane-changing behavior provided in an embodiment of the present application is applied to a data processing end, and the data processing end communicates with a first camera, a fisheye camera, and a second camera. The field of view of the first camera is the oncoming direction of each solid-line road, the field of view of the fisheye camera is the overhead direction of each solid-line road, and the field of view of the second camera is the outgoing direction of each solid-line road. The fisheye camera has overlapping acquisition areas with the first camera and the second camera, respectively, so that when collecting evidence of a vehicle that has changed lanes through a solid line through the first camera, the fisheye camera, and the second camera, there is no blind spot for evidence collection. Specifically, the data processing end can receive each snapshot information reported by each of the first camera and the second camera, and obtain the first virtual queue and the second virtual queue corresponding to each frame image; for each snapshot information received, according to the coordinates of the position frame in the snapshot information and the coordinates of each position frame in the virtual queue corresponding to the camera reporting the snapshot information, the first category ID of the position frame in the snapshot information is associated with the second category ID belonging to the same vehicle, the virtual queue corresponding to the first camera is the first virtual queue, and the virtual queue corresponding to the second camera is the second virtual queue, which is equivalent to, for each snapshot information received and reported by the first camera and the second camera, through the coordinates of the position frame, The first category ID of the captured information is associated with the second category ID of the same vehicle in the corresponding virtual queue reported by the fisheye camera, that is, the position of the same vehicle in the field of view of the oncoming direction of each solid road is associated with the position in the field of view of the overlooking direction, and the position in the field of view of the outgoing direction of each solid road is associated with the position in the field of view of the overlooking direction; thus, in response to receiving the illegal image of any vehicle reported by the fisheye camera, the data processing end can determine the second category ID of the position frame of the vehicle in the illegal image as the target ID, and search for the first category ID associated with the target ID, and generate the evidence result of the vehicle's solid line lane change behavior based on the vehicle image and the illegal image in the captured information to which the first category ID belongs. It can be seen that in this application, the field of view of the fisheye camera is the overlooking direction of each solid road, and the illegal image of any vehicle can be collected, and combined with the vehicle image in the captured information to which the first category ID of the target ID belongs, that is, combined with the captured image of the vehicle in the field of view of the oncoming direction and the captured image in the field of view of the outgoing direction, a complete evidence result about the solid line lane change behavior is generated for the vehicle. This solution can be used to collect evidence from vehicles that change lanes on a solid line.

[0063] The following is an exemplary introduction to the method for collecting evidence of a vehicle's solid line lane changing behavior provided in an embodiment of the present application in conjunction with the accompanying drawings.

[0064] The method for collecting evidence of a vehicle changing lanes on a solid line provided in the embodiment of the present application is applied to a data processing end, such as Figure 1 As shown, the method may include the following steps:

[0065] S101: receiving each captured information reported by each camera of the first camera and the second camera;

[0066] Each captured information reported by each camera includes: a vehicle image captured by the camera when the camera recognizes that a vehicle enters the overlapping acquisition area with the fisheye camera, and the first type ID and coordinates of the position frame of the vehicle in the vehicle image;

[0067] In the evidence collection method for the solid line lane change behavior of a vehicle provided in the embodiment of the present application, the data processing end communicates with the first camera, the fisheye camera and the second camera, the field of view of the first camera is the direction of the vehicle coming from each solid line road, the field of view of the fisheye camera is the top-down direction of each solid line road (it can also be understood that the field of view of the fisheye camera includes the evidence collection blind area of ​​the first camera and the second camera), and the field of view of the second camera is the direction of the vehicle going from each solid line road. The fisheye camera has overlapping acquisition areas with the first camera and the second camera respectively, so that when the first camera, the fisheye camera and the second camera are used to collect evidence for the vehicle that has changed lanes on the solid line, there is no evidence collection blind area, and a complete evidence collection result about the solid line lane change behavior can be generated for the vehicle on the solid line road. Among them, the first camera, the fisheye camera and the second camera can be set up on any lane section containing each solid line road, and the lane section can also contain a traffic intersection, and the specific installation position of the first camera, the fisheye camera and the second camera and the distance between them and the traffic intersection are not limited by the present application.

[0068] For example, Figure 2 The schematic diagram is a diagram of the field of view of the first camera, the fisheye camera and the second camera; it can be seen that the first camera, the fisheye camera and the second camera are installed in sequence from the vehicle's approaching direction to the vehicle's departing direction, the field of view of the first camera is the first field of view, the field of view of the second camera is the second field of view, and the field of view of the fisheye camera is the fisheye field of view. To ensure that there is no blind spot when collecting evidence of the vehicle's solid line lane change behavior on each solid line road through the first camera, the fisheye camera and the second camera, the fisheye field of view of the fisheye camera has an overlapping collection area with the first field of view of the first camera and the second field of view of the second camera; of course, the first camera can capture the vehicle in the overlapping collection area of ​​the first field of view and the fisheye field of view to obtain the captured information of the first camera; the second camera can capture the vehicle in the overlapping collection area of ​​the second field of view and the fisheye field of view to obtain the captured information of the second camera.

[0069] Taking the first camera as a bayonet camera and the second camera as an electric police camera as an example, a schematic diagram of the field of view of the first camera, the fisheye camera and the second camera is as follows: Figure 3 As shown, the checkpoint scene is the field of view of the first camera, and it can be seen that the field of view of the first camera is the oncoming direction of each solid-line road, and can include the starting position of each solid-line road; the electric police scene is the field of view of the second camera, and it can be seen that the field of view of the second camera is the outgoing direction of each solid-line road, and the field of view of the second camera can also include traffic intersections, and the end position of each solid-line road; the fisheye scene is the field of view of the fisheye camera, and the field of view of the fisheye camera is the overhead direction of each solid-line road; and the fisheye camera has overlapping acquisition areas with the first camera and the second camera, respectively, that is, the fisheye scene in the figure has overlapping acquisition areas with the checkpoint scene and the electric police scene, respectively. It should be noted that the time of the first camera, the fisheye camera and the second camera is the same. If they are not the same, time calibration can be performed to make the time of the first camera, the fisheye camera and the second camera the same, and the time can be displayed in the acquisition picture of the field of view of the first camera and the second camera respectively, such as: Friday, May 31, 2024 14:09:40. Of course, the time can also be displayed in the acquisition picture of the field of view of the fisheye camera, and this application does not limit this.

[0070] Each of the first camera and the second camera can capture a vehicle image and identify the first type ID and coordinates of the vehicle's location frame in the vehicle image when it recognizes that a vehicle has entered the overlapping acquisition area of ​​the camera and the fisheye camera. The vehicle image and the first type ID and coordinates can be used as a capture information of the camera; in order to collect evidence of vehicles that have changed lanes on a solid line, each of the first camera and the second camera can report each capture information to the data processing end, and the data processing end can receive each capture information reported by each of the first camera and the second camera. Among them, the method of identifying the first type ID and coordinates of the vehicle's location frame in the vehicle image can be similar to the prior art and will not be repeated here.

[0071] In addition, the way in which each of the first camera and the second camera identifies the presence of a vehicle entering the overlapping acquisition area of ​​the camera and the fisheye camera can be: through the acquisition image of the camera, real-time vehicle area intrusion detection is performed to detect whether the vehicle enters the overlapping acquisition area of ​​the camera and the fisheye camera; of course, any method that can identify the presence of a vehicle entering the overlapping acquisition area of ​​the camera and the fisheye camera is applicable to the present application, and the present application is not limited to this.

[0072] It should also be emphasized that the "first category" in the "first category ID" and the "second category" in the "second category ID" are only used to distinguish the two categories of cameras in terms of naming. One of the two categories of cameras can be considered as the first camera and the second camera, and the other can be considered as the fisheye camera; and, since the first camera and the second camera are considered to belong to the same category, for each camera in the first camera and the second camera, the ID of the location box in its snapshot information can be called the first category ID. Furthermore, the first category ID and coordinates in the snapshot information reported by the first camera are different from the first category ID and coordinates in the snapshot information reported by the second camera. The first category ID in the snapshot information reported by the first camera is the ID set by the first camera according to its corresponding ID setting method, for example: an ID generated in numerical order, and the coordinates in the snapshot information reported by the first camera are the coordinates of the position box in the image coordinate system of the first camera; the first category ID in the snapshot information reported by the second camera is the ID set by the second camera according to its corresponding ID setting method, for example: an ID randomly generated according to numbers and letters, and the coordinates in the snapshot information reported by the second camera are the coordinates of the position box in the image coordinate system of the second camera. Among them, the ID setting methods corresponding to the first camera and the second camera can be any two different ID setting methods, and the first category ID set under the first camera and the first category ID set under the second camera can be different. This application does not limit the ID setting method.

[0073] Of course, each captured information reported by each camera may also include the license plate number of the vehicle in the vehicle image (if the vehicle has a license plate), so as to identify the vehicle by the license plate number, and subsequently associate the evidence collection results of the vehicle's lane change behavior on the solid line with the license plate number.

[0074] S102: Acquire a first virtual queue and a second virtual queue corresponding to each frame image;

[0075] Wherein, each frame of images is collected by the fisheye camera in the process of tracking vehicles entering the field of view, and each frame of images corresponds to a vehicle queue of the second type ID and coordinates of the position box of each vehicle in the image; the first virtual queue corresponding to each frame of images is a vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into coordinates in the image coordinate system of the first camera; the second virtual queue corresponding to each frame of images is a vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into coordinates in the image coordinate system of the second camera;

[0076] Exemplarily, in one implementation, the step of obtaining the first virtual queue and the second virtual queue corresponding to each frame image includes:

[0077] Obtaining a vehicle queue corresponding to each frame of images reported by the fisheye camera, and for the vehicle queue corresponding to each frame of images, converting the coordinates in the vehicle queue corresponding to the frame of images into coordinates in the image coordinate system of the first camera to obtain a first virtual queue corresponding to the frame of images, and converting the coordinates in the vehicle queue corresponding to the frame of images into coordinates in the image coordinate system of the second camera to obtain a second virtual queue corresponding to the frame of images;

[0078] or,

[0079] Receive the first virtual queue and the second virtual queue corresponding to each frame image sent by the fisheye camera.

[0080] In order to collect evidence of vehicles that have changed lanes on a solid line, the fisheye camera can track vehicles that enter the field of view and collect each frame of images during the tracking process. The fisheye camera can identify the second-class ID and coordinates of the position frame of each vehicle in each frame of images, and form a vehicle queue corresponding to the frame of images based on the identified second-class ID and coordinates. For the vehicle queue corresponding to each frame of images, the fisheye camera can convert the coordinates in the vehicle queue corresponding to the image into the coordinates in the image coordinate system of the first camera to form a first virtual queue corresponding to the image; and for the vehicle queue corresponding to each frame of images, convert the coordinates in the vehicle queue corresponding to the image into the coordinates in the image coordinate system of the second camera to form a second virtual queue corresponding to the image, and report the first virtual queue and the second virtual queue corresponding to each frame of images to the data processing end, so that the data processing end uses the first virtual queue and the second virtual queue, as well as each captured information, to associate the same vehicle in the field of view of the first camera, the second camera and the fisheye camera.

[0081] In addition, the fisheye camera can also directly report the vehicle queue corresponding to each frame image (including the second type ID and coordinates of the position box of each vehicle in the image) to the data processing end. For the vehicle queue corresponding to each frame image, the data processing end can convert the coordinates in the vehicle queue corresponding to the image into coordinates in the image coordinate system of the first camera to form a first virtual queue corresponding to the image; and, for the vehicle queue corresponding to each frame image, the data processing end converts the coordinates in the vehicle queue corresponding to the image into coordinates in the image coordinate system of the second camera to form a second virtual queue corresponding to the image; that is, the data processing end can convert the coordinates in the vehicle queue corresponding to each frame image, thereby obtaining the first virtual queue and the second virtual queue corresponding to each frame image.

[0082] It should be noted that, since the second type ID and coordinates of the position frame of each vehicle in each frame image identified by the fisheye camera are required when generating the first virtual queue and the second virtual queue of each frame image, the fisheye camera needs to report the vehicle queue corresponding to each frame image, including the second type ID and coordinates of the position frame of each vehicle in the image (instead of directly reporting each frame image) to the data processing end, so that the data processing end can perform coordinate conversion to generate the first virtual queue and the second virtual queue corresponding to each frame image. It can be seen that in the present application, the fisheye camera or the data processing end can convert the coordinates in the vehicle queue corresponding to each frame image, so as to obtain the first virtual queue and the second virtual queue corresponding to each frame image, and the present application does not limit the device for converting the coordinates in the vehicle queue corresponding to each frame image.

[0083] Specifically, the field of view of the fisheye camera can be calibrated with the field of view of the first camera and the field of view of the second camera, that is, a first mapping relationship between the image coordinate system of the fisheye camera and the image coordinate system of the first camera is established, and a second mapping relationship between the image coordinate system of the fisheye camera and the image coordinate system of the second camera is established. The specific calibration process can be similar to the prior art and will not be described in detail here. The obtained first mapping relationship and the second mapping relationship can be imported into the fisheye camera or the data processing end in the form of a file. It should be noted that the first mapping relationship established is the coordinate mapping relationship between the image coordinate system of the fisheye camera and the image coordinate system of the first camera in the overlapping area of ​​the fisheye camera and the first camera, and the second mapping relationship established is the coordinate mapping relationship between the image coordinate system of the fisheye camera and the image coordinate system of the second camera in the overlapping area of ​​the fisheye camera and the second camera.

[0084] After the fisheye camera captures each frame of image, it can perform image processing on each frame of image to identify the second type ID and coordinates of the position box of each vehicle in each frame of image, so as to form a vehicle queue for each frame of image; then, for the vehicle queue for each frame of image, the fisheye camera or the data processing end uses a first mapping relationship to convert the coordinates in the vehicle queue into coordinates in the image coordinate system of the first camera, so as to form a first virtual queue for the frame of image, and uses a second mapping relationship to convert the coordinates in the vehicle queue into coordinates in the image coordinate system of the second camera, so as to form a second virtual queue for the frame of image.

[0085] It should be noted that each frame image is collected by the fisheye camera during the process of tracking the vehicle within the field of view. It can be each frame image of the continuous video frames collected by the fisheye camera during the tracking process, or each frame image selected from the collected continuous video frames. For example, the continuous video frames collected by the fisheye camera during the tracking process include the 1st, 2nd, 3rd, 4th, 5th...48th, 49th, 50th frame images. The 1st, 2nd, 3rd, 4th, 5th...48th, 49th, 50th frame images can be directly used as each frame image, or alternate frame video images can be selected from the 1st, 2nd, 3rd, 4th, 5th...48th, 49th, 50th frame images, such as: the 1st, 3rd, 5th, 7th...45th, 47th, 49th frame images, or the 2nd, 4th, 6th, 8th...46th, 48th, 50th frame images, as each frame image. This application does not limit this.

[0086] In addition, the coordinates of the position frame of each vehicle in the first virtual queue corresponding to each frame image should belong to the first coordinate range in the image coordinate system of the first camera, and the first coordinate range is the coordinate range corresponding to the overlapping area of ​​the field of view of the first camera and the field of view of the fisheye camera in the image coordinate system of the first camera; the coordinates of the position frame of each vehicle in the second virtual queue corresponding to each frame image should belong to the second coordinate range in the image coordinate system of the second camera, and the second coordinate range is the coordinate range corresponding to the overlapping area of ​​the field of view of the second camera and the field of view of the fisheye camera in the image coordinate system of the second camera. Therefore, the first virtual queue corresponding to each frame image reported by the fisheye camera does not include the second type ID and coordinates of the position frame outside the first coordinate range; the second virtual queue corresponding to each frame image reported by the fisheye camera does not include the second type ID and coordinates of the position frame outside the second coordinate range; it can be understood that before the fisheye camera reports the first virtual queue and the second virtual queue corresponding to each frame image, the second type ID and coordinates of the position frame outside the first coordinate range are removed from the initial first virtual queue, and the second type ID and coordinates of the position frame outside the second coordinate range are removed from the initial second virtual queue, so as to obtain the first virtual queue and the second virtual queue corresponding to each frame image finally uploaded. Alternatively, after the fisheye camera reports the vehicle queue corresponding to each frame image, the data processing end converts the coordinates in the vehicle queue of each frame image to obtain the initial first virtual queue and the initial second virtual queue; then, the second type ID and coordinates of the position frame outside the first coordinate range are removed from the initial first virtual queue, and the second type ID and coordinates of the position frame outside the second coordinate range are removed from the initial second virtual queue, so as to obtain the first virtual queue and the second virtual queue corresponding to each frame image finally.

[0087] S103: for each received snapshot information, according to the coordinates of the location frame in the snapshot information and the coordinates of each location frame in the virtual queue corresponding to the camera reporting the snapshot information, associating the first category ID of the location frame in the snapshot information with the second category ID belonging to the same vehicle;

[0088] The virtual queue corresponding to the first camera is the first virtual queue, and the virtual queue corresponding to the second camera is the second virtual queue;

[0089] After receiving each snapshot information reported by each of the first camera and the second camera, and the first virtual queue and the second virtual queue corresponding to each frame image reported by the fisheye camera, for each snapshot information received, according to the coordinates of the position frame in the snapshot information and the coordinates of each position frame in the virtual queue corresponding to the camera reporting the snapshot information, the first category ID of the position frame in the snapshot information is associated with the second category ID belonging to the same vehicle. That is, for each snapshot information (which may be the snapshot information reported by the first camera or the second camera), according to the coordinates of the position frame in the snapshot information and the coordinates of each position frame in the virtual queue corresponding to the camera reporting the snapshot information (when the camera reporting the snapshot information is the first camera, the corresponding virtual queues are the first virtual queues, and when the camera reporting the snapshot information is the second camera, the corresponding virtual queues are the second virtual queues), the vehicle in the field of view of the camera reporting the snapshot information is associated with the same vehicle in the field of view of the fisheye camera, that is, the vehicle in the field of view of the first camera is associated with the same vehicle in the field of view of the fisheye camera, and the vehicle in the field of view of the second camera is associated with the same vehicle in the field of view of the fisheye camera; thereby, the same vehicle in the field of view of the first camera, the fisheye camera and the second camera is associated through the ID of the position frame. Furthermore, since the present application associates the same vehicle in different camera fields of view through the ID of the position frame, the problem in the related art that a vehicle has no license plate, or the license plate is blocked by the front and rear vehicles, making it impossible to recognize the license plate, or a large vehicle cannot reveal its license plate (such as the rear license plate), making it impossible to collect evidence for a vehicle that changes lanes on a solid line will not occur.

[0090] Exemplarily, in one implementation, each captured information reported by each camera further includes: the acquisition time of the vehicle image; each first virtual queue and second virtual queue is provided with the acquisition time of the corresponding image;

[0091] The method of associating, for each received snapshot information, the first category ID of the location frame in the snapshot information with the second category ID belonging to the same vehicle according to the coordinates of the location frame in the snapshot information and the coordinates of each location frame in the virtual queue corresponding to the camera that reports the snapshot information, includes:

[0092] For each received snapshot information, a virtual queue whose set acquisition time matches the acquisition time in the snapshot information is determined from the virtual queues corresponding to the camera that reported the snapshot information, as the target virtual queue;

[0093] According to the coordinates of the location box in the captured information and the coordinates of each location box in the target virtual queue, the first category ID of the location box in the captured information is associated with the second category ID belonging to the same vehicle in accordance with a predetermined analysis method; wherein the predetermined analysis method is an analysis method that uses the location box overlap ratio as an analysis condition.

[0094] In the present application, each snapshot information reported by each of the first camera and the second camera also includes the acquisition time of the vehicle image, and each first virtual queue and each second virtual queue are set with the corresponding image acquisition time; when the first category ID of the location box summarizing the snapshot information is associated with the second category ID belonging to the same vehicle, for each snapshot information received, the target virtual queue whose set acquisition time matches the acquisition time in the snapshot information can be screened out according to the time from the virtual queue corresponding to the camera reporting the snapshot information, so as to realize the rapid association of the first category ID with the second category ID belonging to the same vehicle.

[0095] Afterwards, the first category ID of the location frame in the captured information can be associated with the second category ID belonging to the same vehicle according to the coordinates of the location frame in the captured information and the coordinates of each location frame in the target virtual queue, in accordance with an analysis method that uses the location frame overlap ratio as an analysis condition.

[0096] Optionally, the associating the first category ID of the location box in the captured information with the second category ID belonging to the same vehicle according to a predetermined analysis method based on the coordinates of the location box in the captured information and the coordinates of each location box in the target virtual queue, includes:

[0097] Calculating the overlap ratio of the location frame in the captured information and the location frames in the target virtual queue according to the coordinates of the location frame in the captured information and the coordinates of the location frames in the target virtual queue to obtain the overlap ratio corresponding to the location frames in the target virtual queue;

[0098] The second category ID of the location box in the target virtual queue whose corresponding overlap ratio is higher than the threshold is associated with the first category ID of the location box in the snapshot information.

[0099] Exemplarily, the overlap ratios of the location frame in the snapshot information and the location frames in the target virtual queue can be calculated based on the coordinates of the location frame in the snapshot information and the coordinates of the location frames in the target virtual queue, so as to obtain the overlap ratios corresponding to the location frames in the target virtual queue; the second category ID of the location frame in the target virtual queue whose corresponding overlap ratio is higher than a threshold is associated with the first category ID of the location frame in the snapshot information. In this case, the location frame in the target virtual queue whose corresponding overlap ratio is higher than the threshold may be a location frame belonging to the same vehicle as the location frame in the snapshot information; alternatively, the second category ID of the location frame in the target virtual queue whose corresponding overlap ratio is the highest may be associated with the first category ID of the location frame in the snapshot information. In this case, the location frame in the target virtual queue whose corresponding overlap ratio is the highest may be a location frame belonging to the same vehicle as the location frame in the snapshot information. This application does not limit this.

[0100] Optionally, for each received snapshot information, determining, from the virtual queues corresponding to the camera that reports the snapshot information, a virtual queue whose set acquisition time meets the time matching relationship with the acquisition time in the snapshot information as the target virtual queue, includes:

[0101] For each snapshot information received, a virtual queue whose difference between a set acquisition time and the acquisition time in the snapshot information is within a predetermined time range is determined from the virtual queues corresponding to the camera reporting the snapshot information as the target virtual queue.

[0102] When determining the target virtual queue from the virtual queues corresponding to the camera that reports the snapshot information based on time, a virtual queue whose difference between the set acquisition time and the acquisition time in the snapshot information is within a predetermined time range can be determined from the virtual queues corresponding to the camera that reports the snapshot information as the target virtual queue (or, if there are multiple virtual queues whose difference between the set acquisition time and the acquisition time in the snapshot information is within the predetermined time range, the virtual queue whose difference between the set acquisition time and the acquisition time in the snapshot information is the smallest can be used as the target virtual queue); for example: the predetermined time range can be 500ms or 5min, etc., and can be configured according to needs or actual conditions, and this application does not limit this.

[0103] It should be noted that, for each snapshot information reported by the first camera or the second camera, the first category ID of the location box in the snapshot information can be associated with the second category ID belonging to the same vehicle in the above-mentioned manner. The difference lies in that the first category ID of the location box in the snapshot information reported by the first camera is associated with the second category ID of the location box in the first virtual queue belonging to the same vehicle, and the first category ID of the location box in the snapshot information reported by the second camera is associated with the second category ID of the location box in the second virtual queue belonging to the same vehicle.

[0104] S104: In response to receiving the illegal image of any vehicle reported by the fisheye camera, determining the second category ID of the position box of the vehicle in the illegal image as the target ID, searching for the first category ID associated with the target ID, and generating the evidence result of the vehicle's lane change behavior on the solid line based on the illegal image and the vehicle image in the captured information to which the found first category ID belongs;

[0105] The illegal image is an image captured by the fisheye camera when it identifies any vehicle making an illegal lane change.

[0106] After associating the same vehicle in the field of view of the first camera, the fisheye camera and the second camera, evidence can be collected for the solid line lane change behavior of the vehicle (or, in other words, if the evidence collection results already exist, the evidence collection results are associated with the same vehicle); the field of view of the fisheye camera is the top-down direction of the solid line road, and the vehicle usually changes lanes in the field of view of the fisheye camera. In the process of tracking the vehicle, the fisheye camera can identify whether the vehicle changes lanes in the solid line, and collect the illegal image of the vehicle and report it to the data processing end; the data processing end receives the illegal image of any vehicle reported by the fisheye camera, and can determine the second type ID of the position box of the vehicle in the illegal image as the target ID, and through the above steps The first category ID and the second category ID associated with the step are searched for the first category ID associated with the target ID, and based on the illegal image and the vehicle image in the captured information to which the found first category ID belongs (including the vehicle image for the first category ID captured by the first camera and the vehicle image for the first category ID captured by the second camera), the forensic result of the vehicle's lane changing behavior at a solid line is generated; exemplarily, the illegal image and the vehicle image in the captured information described in the found first category ID can be directly used as the forensic result of the vehicle's lane changing behavior, or the illegal image and the vehicle image in the captured information described in the found first category ID can be spliced ​​to obtain the forensic result of the vehicle's lane changing behavior at a solid line.

[0107] Exemplarily, in one implementation, the illegal image of any vehicle includes: an image of the vehicle before changing lanes in a solid line, an image of the vehicle when changing lanes in a solid line, and an image of the vehicle after changing lanes in a solid line;

[0108] The fisheye camera collects illegal images of any vehicle in the following ways:

[0109] After recognizing that the vehicle has entered a solid-line road, an image of the vehicle is collected to obtain an image of the vehicle before the solid-line lane change;

[0110] When the vehicle is identified as being located on a lane line of a solid line road, an image of the vehicle is captured to obtain an image of the vehicle when changing lanes along a solid line;

[0111] When it is identified that the lane number of the solid lane where the vehicle is located is different from the lane number when entering the solid road, an image of the vehicle is captured to obtain an image of the vehicle after the solid lane change.

[0112] The illegal image of any vehicle includes the image of the vehicle before the solid line lane change, the image of the vehicle when the solid line lane change and the image of the vehicle after the solid line lane change, so as to clearly characterize the process of the vehicle's solid line lane change behavior; when the fisheye camera collects illegal images for any vehicle, after identifying that the vehicle enters the solid line road, it collects images of the vehicle to obtain the image of the vehicle before the solid line lane change; when identifying that the vehicle is located on the lane line of the solid line road, it collects images of the vehicle to obtain the image of the vehicle when the solid line lane change is obtained; when identifying that the lane number of the solid line lane where the vehicle is located is different from the lane number when entering the solid line road, it collects images of the vehicle to obtain the image of the vehicle after the solid line lane change. Among them, regardless of whether the vehicle has a license plate, since the field of view of the fisheye camera is the overhead direction of each solid-line road, the illegal image of any vehicle collected by the fisheye camera (the image of the vehicle before the solid-line lane change, the image of the vehicle when the solid-line lane change, and the image of the vehicle after the solid-line lane change) does not include the license plate number of the vehicle; and the captured information reported by each camera of the first camera and the second camera includes the vehicle image. After associating the first type ID and the second type ID belonging to the same vehicle, the vehicle information of the vehicle (such as: license plate number, vehicle model, vehicle color, etc., if the license plate number exists, the license plate number can be directly used to identify the vehicle) can be determined based on the vehicle image, thereby associating the vehicle information of the vehicle with the illegal image of the vehicle collected by the fisheye camera.

[0113] It should be noted that the fisheye camera can use its collected images to perform real-time area intrusion detection of vehicles, so as to detect whether the vehicle has entered a solid road, whether it is located on the lane line of a solid road, and whether the lane number of the solid lane where the vehicle is located is different from the lane number when entering the solid road; wherein, the area within the field of view of the fisheye camera can be configured, for example: the field of view of the fisheye camera can be configured as a solid lane area and a non-solid lane area, and the lane area can be configured as different solid lanes according to the lane numbers of each lane, thereby realizing the above-mentioned detection.

[0114] In the technical solution of the present application, the operations involved in acquiring, storing, using, processing, transmitting, providing and disclosing the captured information, frame images, virtual queues, vehicle queues and illegal images are all carried out with the user's authorization.

[0115] The method for collecting evidence of a vehicle's solid-line lane-changing behavior provided in an embodiment of the present application is applied to a data processing end, and the data processing end communicates with a first camera, a fisheye camera, and a second camera. The field of view of the first camera is the oncoming direction of each solid-line road, the field of view of the fisheye camera is the overhead direction of each solid-line road, and the field of view of the second camera is the outgoing direction of each solid-line road. The fisheye camera has overlapping acquisition areas with the first camera and the second camera, respectively, so that when collecting evidence of a vehicle that has changed lanes through a solid line through the first camera, the fisheye camera, and the second camera, there is no blind spot for evidence collection. Specifically, the data processing end can receive each snapshot information reported by each of the first camera and the second camera, and obtain the first virtual queue and the second virtual queue corresponding to each frame image; for each snapshot information received, according to the coordinates of the position frame in the snapshot information and the coordinates of each position frame in the virtual queue corresponding to the camera reporting the snapshot information, the first category ID of the position frame in the snapshot information is associated with the second category ID belonging to the same vehicle, the virtual queue corresponding to the first camera is the first virtual queue, and the virtual queue corresponding to the second camera is the second virtual queue, which is equivalent to, for each snapshot information received and reported by the first camera and the second camera, through the coordinates of the position frame, The first category ID of the captured information is associated with the second category ID of the same vehicle in the corresponding virtual queue reported by the fisheye camera, that is, the position of the same vehicle in the field of view of the oncoming direction of each solid road is associated with the position in the field of view of the overlooking direction, and the position in the field of view of the outgoing direction of each solid road is associated with the position in the field of view of the overlooking direction; thus, in response to receiving the illegal image of any vehicle reported by the fisheye camera, the data processing end can determine the second category ID of the position frame of the vehicle in the illegal image as the target ID, and search for the first category ID associated with the target ID, and generate the evidence result of the vehicle's solid line lane change behavior based on the vehicle image and the illegal image in the captured information to which the first category ID belongs. It can be seen that in this application, the field of view of the fisheye camera is the overlooking direction of each solid road, and the illegal image of any vehicle can be collected, and combined with the vehicle image in the captured information to which the first category ID of the target ID belongs, that is, combined with the captured image of the vehicle in the field of view of the oncoming direction and the captured image in the field of view of the outgoing direction, a complete evidence result about the solid line lane change behavior is generated for the vehicle. This solution can be used to collect evidence from vehicles that change lanes on a solid line.

[0116] Exemplarily, in another embodiment of the present application, before generating the evidence result of the vehicle's lane change behavior on a solid line based on the illegal image and the vehicle image in the captured information to which the first category ID belongs, the method further includes:

[0117] Acquire a target video clip captured by the fisheye camera; wherein the target video clip is a video clip whose capture time range includes the capture time of the illegal image and has a predetermined duration;

[0118] The generating of the evidence collection result of the vehicle's lane changing behavior on a solid line based on the illegal image and the vehicle image in the captured information to which the first category ID belongs includes:

[0119] Determine the vehicle information of the vehicle based on the vehicle image in the captured information to which the determined first category ID belongs; wherein the vehicle information of the vehicle is used to identify the vehicle;

[0120] The vehicle information of the vehicle, the vehicle image in the captured information to which the first category ID is found, the illegal image and the target video clip are associated to obtain the evidence collection result of the vehicle's lane changing behavior on the solid line.

[0121] In an embodiment of the present application, the evidence collection results of the vehicle's solid line lane changing behavior may also include a video clip of the vehicle's solid line lane changing. The target video clip may be acquired first, which is captured by a fisheye camera and has a predetermined duration, and a capture time range that includes the capture time of the illegal image. The predetermined duration may be the duration represented by the time difference between the image of the vehicle before the solid line lane changing and the image of the vehicle after the solid line lane changing.

[0122] When generating the forensic results of the vehicle's lane changing behavior on a solid line, the vehicle information used to identify the vehicle, such as license plate number, vehicle model, vehicle color, etc., can be determined based on the vehicle image in the captured information described in the determined first category ID; and the vehicle information of the vehicle, the vehicle image in the captured information belonging to the found first category ID, the illegal image and the target video clip are associated to obtain the forensic results of the vehicle's lane changing behavior on a solid line.

[0123] It can be seen that in the embodiment of the present application, it is also possible to determine the target video clip of the vehicle violation and determine the vehicle information that identifies the vehicle. The vehicle information of the vehicle can be associated with the vehicle image, the violation image and the target video clip in the captured information belonging to the first category ID found, so as to better characterize the process of the vehicle's solid line lane change behavior and obtain more comprehensive evidence collection results on the solid line lane change behavior.

[0124] Optionally, in another embodiment of the present application, generating a forensic result of the vehicle's lane changing behavior on a solid line based on the illegal image and the vehicle image in the captured information to which the first category ID belongs, includes:

[0125] Based on the vehicle image in the captured information to which the first category ID belongs, a close-up image of the vehicle in the oncoming direction and a close-up image of the vehicle in the outgoing direction are generated;

[0126] Determine the vehicle information of the vehicle based on the vehicle image in the captured information to which the determined first category ID belongs; wherein the vehicle information of the vehicle is used to identify the vehicle;

[0127] The vehicle image in the captured information belonging to the first category ID found, the illegal image, the close-up image of the vehicle in the oncoming direction, and the close-up image of the vehicle in the outgoing direction are synthesized to obtain a synthesized image, and the obtained synthesized image is associated with the determined vehicle information to obtain the evidence collection result of the vehicle's lane changing behavior on the solid line.

[0128] When generating the forensic results of the vehicle's lane changing behavior on a solid line, the present application can also generate a close-up image of the vehicle's oncoming direction and a close-up image of the vehicle's outgoing direction based on the vehicle image in the captured information belonging to the first category ID found (the vehicle image in the captured information of the first camera and the vehicle image in the captured information of the second camera). For example, the vehicle image of the vehicle's oncoming direction (the vehicle image in the captured information of the first camera) is enlarged, and the image containing the vehicle is captured to obtain a close-up image of the vehicle's oncoming direction, and in a similar manner, a close-up image of the vehicle's outgoing direction is obtained.

[0129] When generating the forensic results of the vehicle's lane changing behavior on a solid line, the vehicle information used to identify the vehicle, such as license plate number, vehicle model, vehicle color, etc., can be determined based on the vehicle image in the captured information described in the determined first category ID; and the vehicle image, violation image, close-up image of the vehicle's oncoming direction, and close-up image of the vehicle's outgoing direction in the captured information belonging to the found first category ID are synthesized (for example, image stitching is performed in a predetermined position and order, the specific position and order can be set as required, and are not limited in this application) to obtain a synthesized image, and the obtained synthesized image is associated with the determined vehicle information to obtain the forensic results of the vehicle's lane changing behavior on a solid line.

[0130] It can be seen that in the embodiment of the present application, a close-up image of the vehicle's oncoming direction and a close-up image of the vehicle's outgoing direction can also be generated, and the vehicle information identifying the vehicle can be determined. The vehicle image, the violation image, the close-up image of the vehicle's oncoming direction and the close-up image of the vehicle's outgoing direction in the captured information belonging to the first category ID found can be synthesized to obtain a composite image. The composite image can better characterize the process of the vehicle's solid line lane change behavior. The obtained composite image can be associated with the determined vehicle information to obtain a more comprehensive evidence collection result on the solid line lane change behavior.

[0131] Among them, it should be noted that, for any vehicle, the target video clips and illegal images of the vehicle collected by the fisheye camera, the vehicle image in the captured information described in the first category ID found based on the illegal image, the close-up image of the vehicle in the oncoming direction and the close-up image of the vehicle in the outgoing direction and other data can all be stored in the data processing end (specifically, they can be stored in the memory of the data processing end) to generate the evidence results of the vehicle's lane change behavior on the solid line.

[0132] The present application also provides a system for collecting evidence of a vehicle's lane change behavior. Figure 4 As shown, it includes: a data processing end 410, a first camera 420, a second camera 430 and a fisheye camera 440; the data processing end communicates with the first camera, the fisheye camera and the second camera, the field of view of the first camera is the direction of the vehicle coming from each solid-line road, the field of view of the fisheye camera is the direction of looking down at each solid-line road, the field of view of the second camera is the direction of the vehicle going out of each solid-line road, and the fisheye camera has overlapping acquisition areas with the first camera and the second camera respectively;

[0133] Each of the first camera 420 and the second camera 430 is used to capture the vehicle when recognizing that a vehicle enters the overlapping collection area with the fisheye camera, and report the captured information obtained by capturing the vehicle to the data processing end; wherein the captured information includes a captured vehicle image, a first type ID and coordinates of a location frame of the vehicle in the vehicle image;

[0134] The fisheye camera 440 is used to track vehicles entering the field of view, and perform image analysis and processing on each frame of images collected during the tracking process to obtain a vehicle queue of the second type ID and coordinates of a position frame of each vehicle in the image corresponding to each frame of images;

[0135] The data processing end 410 is used to receive each snapshot information reported by each of the first camera and the second camera, and obtain the first virtual queue and the second virtual queue corresponding to each frame image; and, for each snapshot information received, according to the coordinates of the position frame in the snapshot information and the coordinates of each position frame in the virtual queue corresponding to the camera reporting the snapshot information, associate the first category ID of the position frame in the snapshot information with the second category ID belonging to the same vehicle; wherein the virtual queue corresponding to the first camera is the first virtual queue, and the virtual queue corresponding to the second camera is the second virtual queue; the first virtual queue corresponding to each frame image is the vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into the coordinates in the image coordinate system of the first camera; the second virtual queue corresponding to each frame image is the vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into the coordinates in the image coordinate system of the second camera;

[0136] The fisheye camera 440 is also used to collect images when any vehicle is identified to have made an illegal lane change, obtain an illegal image of the vehicle, and report the illegal image to the data processing end;

[0137] The data processing end 410 is also used to receive the violation image, determine the second category ID of the position box of the vehicle in the violation image as the target ID, search for the first category ID associated with the target ID, and generate the evidence results of the vehicle's lane change behavior on the solid line based on the violation image and the vehicle image in the captured information to which the found first category ID belongs.

[0138] The embodiment of the present application provides an evidence collection system for a vehicle's solid line lane change behavior, comprising: a data processing end, a first camera, a second camera and a fisheye camera, the data processing end communicates with the first camera, the fisheye camera and the second camera, the field of view of the first camera is the oncoming direction of each solid line road, the field of view of the fisheye camera is the top-down direction of each solid line road, the field of view of the second camera is the outgoing direction of each solid line road, and the fisheye camera has overlapping acquisition areas with the first camera and the second camera respectively, so that when collecting evidence of a vehicle that has changed lanes through a solid line through the first camera, the fisheye camera and the second camera, there is no blind spot for evidence collection. Specifically, the data processing end can receive each snapshot information reported by each of the first camera and the second camera, and obtain the first virtual queue and the second virtual queue corresponding to each frame image; for each snapshot information received, according to the coordinates of the position frame in the snapshot information and the coordinates of each position frame in the virtual queue corresponding to the camera reporting the snapshot information, the first category ID of the position frame in the snapshot information is associated with the second category ID belonging to the same vehicle, the virtual queue corresponding to the first camera is the first virtual queue, and the virtual queue corresponding to the second camera is the second virtual queue, which is equivalent to, for each snapshot information received and reported by the first camera and the second camera, through the coordinates of the position frame, The first category ID of the captured information is associated with the second category ID of the same vehicle in the corresponding virtual queue reported by the fisheye camera, that is, the position of the same vehicle in the field of view of the oncoming direction of each solid road is associated with the position in the field of view of the overlooking direction, and the position in the field of view of the outgoing direction of each solid road is associated with the position in the field of view of the overlooking direction; thus, in response to receiving the illegal image of any vehicle reported by the fisheye camera, the data processing end can determine the second category ID of the position frame of the vehicle in the illegal image as the target ID, and search for the first category ID associated with the target ID, and generate the evidence result of the vehicle's solid line lane change behavior based on the vehicle image and the illegal image in the captured information to which the first category ID belongs. It can be seen that in this application, the field of view of the fisheye camera is the overlooking direction of each solid road, and the illegal image of any vehicle can be collected, and combined with the vehicle image in the captured information to which the first category ID of the target ID belongs, that is, combined with the captured image of the vehicle in the field of view of the oncoming direction and the captured image in the field of view of the outgoing direction, a complete evidence result about the solid line lane change behavior is generated for the vehicle. This solution can be used to collect evidence from vehicles that change lanes on a solid line.

[0139] The following is a detailed introduction to the vehicle solid line lane changing behavior evidence collection system of the present application in conjunction with a specific embodiment.

[0140] The evidence collection system for the solid-line lane-changing behavior of vehicles in this application consists of two parts: a front-end camera and a data processing end (for example only, not limited). The front-end camera includes a bayonet camera (first camera), an electric police camera (second camera) and a fisheye camera. The bayonet camera captures the front image of the vehicle, the electric police camera captures the rear (direction) image of the vehicle, and the fisheye camera captures three images of the illegal lane-changing process in the blind spot (the image before the vehicle changes lanes, the image when the vehicle changes lanes, and the image after the vehicle changes lanes); the data processing end is responsible for collecting video, receiving and storing captured data, matching the targets of the three channels of bayonet, electric police, and fisheye, and after the match is successful, synthesizing multiple illegal captured images of the same vehicle, etc., to generate the evidence collection results of the vehicle's solid-line lane-changing behavior.

[0141] like Figure 2 As shown, the bayonet camera (first camera), the electric police camera (second camera) and the fisheye camera are mounted on a pole with a height of 6 meters. The shooting direction of the bayonet camera (first field of view range) is the scene facing forward of the vehicle (the direction of oncoming vehicles of each solid-line road), the shooting direction of the electric police camera (second field of view range) is the scene facing backward of the vehicle (the direction of outgoing vehicles of each solid-line road), and the shooting direction of the fisheye camera (fisheye field of view range) is the top-down perspective of the solid-line roads (the top-down direction of each solid-line road), which is used to shoot above the vehicle. There is a 50-meter blind spot between the checkpoint camera and the electric police camera (i.e., between the first field of view and the second field of view). If a vehicle changes lanes in the blind spot, the checkpoint camera and the electric police camera will not be able to capture the image of the vehicle's illegal lane change. The fisheye camera supplements this blind spot. The fisheye field of view is 80 meters, and there is a 15-meter overlapping area with the first field of view of the checkpoint camera and the second field of view of the electric police camera. In the overlapping area, the vehicle's ID can be tracked and relayed to determine the corresponding ID of the same vehicle in the three videos (images or videos captured by the three cameras respectively), so as to match and synthesize the captured images of the illegal vehicle and complete the evidence collection process on the vehicle's solid line lane change behavior.

[0142] The interactive process of the vehicle solid line lane change behavior evidence collection system of this application is as follows Figure 5 As shown:

[0143] 1. Deploy the bayonet camera, electric police camera, and fisheye camera to the data processing end (i.e. establish the communication relationship between the bayonet camera, electric police camera, fisheye camera and the data processing end), and the three front-end cameras support synchronous time calibration, and the system time is consistent;

[0144] 2. When the vehicle arrives at the checkpoint capture position, the checkpoint camera captures the front image of the vehicle and executes S501: sending the checkpoint capture information to the data processing end. The checkpoint capture information includes: the front image of the vehicle, and the vehicle position frame ID, position frame coordinates, capture time, lane number, license plate (if there is no license plate in the image, there may be no license plate in the checkpoint capture information), vehicle type, etc. obtained by identifying the front image of the vehicle. The checkpoint capture position is located in the overlapping area of ​​the first field of view and the fisheye field of view.

[0145] 3. The fisheye camera maps the vehicle queue of each frame image collected within the fisheye field of view to the checkpoint to obtain a first virtual queue, and executes step S502: sending the first virtual queue mapped by the fisheye to the checkpoint to the data processing end. The process of mapping the vehicle queue of each frame image of the fisheye to the checkpoint and reporting it to the data processing end is as follows: Figure 6 As shown, firstly, the fisheye field of view range and the bayonet field of view range are calibrated (for example, by calibrating the fisheye scene image and the bayonet scene image of the overlapping area of ​​the fisheye field of view range and the bayonet field of view range at the same time), and the calibration generates a mapping model file about the image coordinates of the fisheye camera to the image coordinates of the bayonet camera, and the mapping model file is imported into the fisheye camera, that is, S601: importing the mapping model, into the fisheye camera, at this time, the mapping model is equivalent to existing in the fisheye camera; then, S602: acquiring each frame image of the fisheye, and each frame image of the fisheye (each frame image of the continuous video frames collected by the fisheye camera during the tracking process, or, from the collected continuous video frames) The obtained frames are selected and sent to the algorithm for S603: image processing. The algorithm processing frame rate can be 12.5fps. The algorithm outputs the vehicle queue (information includes video frame time, target frame coordinates, fisheye target ID, etc.), that is, S604: generating a fisheye vehicle queue (each frame image corresponds to a vehicle queue), sending the vehicle queue to the mapping model for conversion, and then S605: generating a first virtual queue mapped to the checkpoint, and executing S606: filtering abnormal data (the location frame and ID of the vehicle whose converted coordinates exceed the image coordinate range of the checkpoint camera, etc.), and then executing S607: uploading the first virtual queue after the abnormal data is filtered to the data processing end for subsequent matching of the same vehicle.

[0146] 4. After receiving the checkpoint snapshot information and the first virtual queue mapped to the checkpoint by the fisheye, the data processing end performs step S503: vehicle matching with the checkpoint snapshot information and the vehicle location frame in the first virtual queue, and associates the first category ID of the location frame in the checkpoint snapshot information with the second category ID belonging to the same vehicle. The specific association process is as follows Figure 7As shown, first, the data processing end executes step S701: receiving the virtual queue mapped from the fisheye to the checkpoint or the electric police (the virtual queue mapped from the fisheye to the checkpoint is the first virtual queue, and the virtual queue mapped from the fisheye to the electric police is the second virtual queue, and the subsequent association between the fisheye camera and the electric police camera for the same vehicle is similar thereto), obtaining the checkpoint virtual queue or the electric police virtual queue, and caching the virtual queue into the database, which will save virtual queues within 5 minutes (configurable and not limited); the data processing end continues to execute step S702: receiving the checkpoint or electric police snapshot information, and then executes step S703: judging whether a match is required according to the configuration (for example, whether image synthesis forensics of solid line lane change behavior is configured), if not, ending directly; if yes, searching for a virtual queue within 500ms (configurable as needed and not limited) of the time difference according to the checkpoint snapshot time, and executing step S704: judging whether Whether the target virtual queue is found (the virtual queue in which the difference between the set acquisition time and the acquisition time in the captured information is within the predetermined time range, for example: the virtual queue with the smallest time difference within 500ms), if no, then end directly; if yes, execute step S705: calculate the overlap ratio between the position box in the captured information and the position box in the target virtual queue, and execute step S706: determine whether the overlap ratio is greater than the threshold; if no (i.e. not greater than the threshold), the matching fails and ends directly; if yes (i.e. greater than the threshold), execute step S707: the matching is successful, and the checkpoint or electric police captured image is bound to the second type ID, that is, the first type ID of the position box in the checkpoint captured information is associated with the second type ID belonging to the same vehicle in the target virtual queue to obtain the associated checkpoint captured information or electric police captured information, and finally save the checkpoint captured information or electric police captured information to the database for subsequent illegal lane change image synthesis.

[0147] 5. The fisheye camera executes step S504: capturing illegal lane changes in the blind spot. After the vehicle enters the solid lane, the fisheye camera captures the first illegal lane change (the image before the vehicle changes lanes in the solid lane). When the vehicle is on the solid lane, the second illegal lane change is captured (the image during the vehicle changing lanes in the solid lane). When the vehicle's lane number is changed from the first lane number (or the lane number of the solid lane where the vehicle is located is different from the lane number when entering the solid lane), the third illegal lane change is captured (the image after the vehicle changes lanes in the solid lane). The fisheye camera captures the illegal lane change images as shown in the figure. Fig. 9 As shown in the figure, from left to right, they are the first, second and third images captured by the fisheye camera (i.e., images before, during and after the vehicle changes lanes; it should be noted that due to the characteristics of the fisheye camera, the captured images are not necessarily planar images, but may be images with convex vision effects; Fig. 9The fisheye camera sends the three illegal lane change capture image information (the information includes the capture time, the ID of the location box, the lane number, etc.) to the data processing end, that is, executes step S505: sending the illegal image to the data processing end.

[0148] 6. The fisheye camera maps the vehicle queue of each frame image collected within the fisheye field of view to the electric police, obtains a second virtual queue, and executes step S506: sends the second virtual queue mapped from the fisheye to the electric police to the data processing end. The mapping process is similar to step 3 and will not be repeated here.

[0149] 7. When the vehicle arrives at the electric police capture position, the electric police camera captures the image of the vehicle's back and executes step S507: sending the electric police capture information to the data processing end. The electric police capture information includes: the vehicle's back image, and the vehicle position frame ID, position frame coordinates, capture time, lane number, license plate (if there is no license plate in the image, there may be no license plate in the checkpoint capture information), vehicle type, etc. obtained by identifying the vehicle's back image. The electric police capture position is located in the overlapping area of ​​the second field of view and the fisheye field of view.

[0150] 8. After the data processing end receives the police capture information and the fisheye mapping to the second virtual queue of the police, the police capture information and the vehicle location frame in the second virtual queue are executed in step S508: vehicle matching, which associates the first category ID of the location frame in the police capture information with the second category ID belonging to the same vehicle. The specific association process is similar to step 4 and will not be repeated here.

[0151] 9. After the data processing end deploys the front-end equipment, it will receive the code stream information of the front-end camera, that is, the checkpoint camera executes step S509: sending the checkpoint code stream information to the data processing end, the fisheye camera executes step S510: sending the fisheye code stream information to the data processing end, and the electric police camera executes step S511: sending the electric police code stream information to the data processing end; the data processing end receives the code stream and stores the video of the code stream to the hard disk in real time for capturing the video of the illegal process of changing lanes on a solid line; and the data processing end executes step S512: synthesizing multiple pictures and capturing illegal video clips.

[0152] 10. The process of image synthesis and forensic results generation at the data processing end is as follows Figure 8As shown, the data processing end first executes step S801: receiving the fisheye illegal image, and then executes step S802: searching for the checkpoint and electric police snapshot information according to the second category ID (i.e. searching for the checkpoint snapshot information and electric police snapshot information corresponding to the first category ID according to the associated ID from the stored checkpoint snapshot information and electric police snapshot information), and executes step S803: judging whether the corresponding checkpoint and electric police snapshot information are found, and if the judgment result is no, then the process ends directly; if the judgment result is yes, then execute step S804: performing image synthesis, i.e. performing image synthesis based on the checkpoint snapshot information (checkpoint snapshot image), electric police snapshot information (electric police snapshot image) and the fisheye illegal image; and, based on the fisheye video database (a database storing the video corresponding to the code stream sent by the fisheye camera), execute step S805: capturing the video clip according to the fisheye illegal lane change event, and step S806: combining to generate a complete illegal lane change evidence collection result. Among them, the synthetic diagram for image synthesis is shown in FIG. Fig.10 As shown, it consists of 3 fisheye lane change images (the first fisheye illegal lane change image, the second fisheye illegal lane change image, and the third fisheye illegal lane change image) + 1 checkpoint front-facing snapshot image + 1 checkpoint close-up image + 1 electric police back-facing snapshot image + 1 electric police close-up image.

[0153] In addition, it should be noted that when collecting evidence of a vehicle's solid line lane change behavior, it is not necessary to collect evidence of the vehicle's solid line lane change behavior immediately after receiving a captured image or a virtual queue. Instead, evidence of the vehicle's solid line lane change behavior can be collected after a certain period of time to obtain a complete evidence collection result. This application does not limit this.

[0154] In view of the blind spots of checkpoints and electric police cameras, the present application adds a fisheye camera. The fisheye camera can capture illegal images of the actual illegal lane changing process of the vehicle in the blind spots of the checkpoints and electric police cameras, and capture images of the vehicle before, during and after the lane change, and provide fisheye illegal video clips; the present application uses a vehicle matching method that maps the coordinates of the overlapping area between the field of view of the fisheye camera and the field of view of the checkpoint and electric police camera to obtain multi-scene captured images of the vehicle at the checkpoint, fisheye and electric police, and associates the same vehicle through the ID of the position frame to solve the problem that the license plate cannot be obtained in a single fisheye scene; the present application does not rely on the license plate number in the captured images of the checkpoint and electric police for matching, but associates the same vehicle through the ID of the position frame, and can identify illegal lane changing behaviors in scenarios such as no license plate, obstructed front and rear license plates of the vehicle, or a large vehicle that is too long to expose the rear of the vehicle and cannot identify the license plate in the field of view of the electric police, and generate evidence collection results for the vehicle's solid line lane changing behavior.

[0155] Based on the above method embodiment, the present application also provides a vehicle solid line lane change behavior evidence collection device, which is applied to a data processing end, and the data processing end communicates with a first camera, a fisheye camera and a second camera. The field of view of the first camera is the direction of the oncoming vehicle of each solid line road, the field of view of the fisheye camera is the top-down direction of each solid line road, and the field of view of the second camera is the outgoing direction of each solid line road. The fisheye camera has overlapping collection areas with the first camera and the second camera respectively; Fig.11 As shown, the device comprises:

[0156] The first receiving module 1110 is used to receive each captured information reported by each camera of the first camera and the second camera; wherein each captured information reported by each camera includes: a vehicle image captured by the camera when the camera recognizes that a vehicle enters the overlapping acquisition area with the fisheye camera, and a first type ID and coordinates of a position frame of the vehicle in the vehicle image;

[0157] The acquisition module 1120 is used to acquire the first virtual queue and the second virtual queue corresponding to each frame image; wherein each frame image is collected by the fisheye camera in the process of tracking the vehicles entering the field of view, and each frame image in each frame image corresponds to a vehicle queue of the second type ID and coordinates of the position frame of each vehicle in the image; the first virtual queue corresponding to each frame image is a vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into the coordinates in the image coordinate system of the first camera; the second virtual queue corresponding to each frame image is a vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into the coordinates in the image coordinate system of the second camera;

[0158] The association module 1130 is used to associate, for each received snapshot information, a first type ID of the location frame in the snapshot information with a second type ID belonging to the same vehicle according to the coordinates of the location frame in the snapshot information and the coordinates of each location frame in the virtual queue corresponding to the camera reporting the snapshot information; wherein the virtual queue corresponding to the first camera is the first virtual queue, and the virtual queue corresponding to the second camera is the second virtual queue;

[0159] The generation module 1140 is used to, in response to receiving an illegal image of any vehicle reported by the fisheye camera, determine the second category ID of the position box of the vehicle in the illegal image as the target ID, search for the first category ID associated with the target ID, and generate the evidence result of the vehicle's solid line lane changing behavior based on the illegal image and the vehicle image in the captured information to which the found first category ID belongs; wherein the illegal image is the image captured by the fisheye camera when it identifies that any vehicle has committed an illegal lane changing behavior.

[0160] The embodiment of the present application further provides a data processing device, the data processing device communicates with a first camera, a fisheye camera and a second camera, the field of view of the first camera is the direction of the vehicle coming from each solid-line road, the field of view of the fisheye camera is the direction of looking down at each solid-line road, the field of view of the second camera is the direction of the vehicle going out of each solid-line road, and the fisheye camera has overlapping acquisition areas with the first camera and the second camera respectively; Fig.12 As shown, the data processing device includes:

[0161] Memory 1201, used for storing computer programs;

[0162] The processor 1202 is used to implement any of the above-mentioned methods for collecting evidence of a vehicle changing lanes on a solid line when executing the program stored in the memory 1201.

[0163] Furthermore, the above-mentioned data processing device may further include a communication bus and / or a communication interface, and the processor 1202, the communication interface, and the memory 1201 communicate with each other via the communication bus.

[0164] The communication bus mentioned in the above data processing 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.

[0165] The communication interface is used for communication between the above data processing device and other devices.

[0166] 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.

[0167] 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.

[0168] In another embodiment of the present application, an all-in-one device is provided, comprising a first camera, a fisheye camera, a second camera and a data processing end; the data processing end communicates with the first camera, the fisheye camera and the second camera, the field of view of the first camera is the direction of the vehicle coming from each solid-line road, the field of view of the fisheye camera is the direction of looking down at each solid-line road, the field of view of the second camera is the direction of the vehicle going out of each solid-line road, and the fisheye camera has overlapping acquisition areas with the first camera and the second camera respectively;

[0169] Each of the first camera and the second camera is used to capture the vehicle when recognizing that a vehicle enters the overlapping collection area with the fisheye camera, and report the captured information obtained by capturing the vehicle to the data processing end; wherein the captured information includes a captured vehicle image, a first type ID and coordinates of a position frame of the vehicle in the vehicle image;

[0170] The fisheye camera is used to track vehicles entering the field of view, and perform image analysis and processing on each frame of images collected during the tracking process to obtain a vehicle queue of the second type ID and coordinates of a position frame of each vehicle in the image corresponding to each frame of images;

[0171] The data processing end is used to receive each snapshot information reported by each of the first camera and the second camera, and obtain the first virtual queue and the second virtual queue corresponding to each frame image; and, for each snapshot information received, according to the coordinates of the position frame in the snapshot information and the coordinates of each position frame in the virtual queue corresponding to the camera reporting the snapshot information, associate the first category ID of the position frame in the snapshot information with the second category ID belonging to the same vehicle; wherein the virtual queue corresponding to the first camera is the first virtual queue, and the virtual queue corresponding to the second camera is the second virtual queue; the first virtual queue corresponding to each frame image is the vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into the coordinates in the image coordinate system of the first camera; the second virtual queue corresponding to each frame image is the vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into the coordinates in the image coordinate system of the second camera;

[0172] The fisheye camera is further used to collect images when identifying any vehicle that has made an illegal lane change, obtain an illegal image of the vehicle, and report the illegal image to the data processing end;

[0173] The data processing end is also used to receive the violation image, determine the second category ID of the position box of the vehicle in the violation image as the target ID, search for the first category ID associated with the target ID, and generate the evidence results of the vehicle's lane changing behavior on the solid line based on the violation image and the vehicle image in the captured information to which the found first category ID belongs.

[0174] 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 the method for collecting evidence of any of the above-mentioned vehicle solid line lane changing behaviors are implemented.

[0175] In another embodiment provided in the present application, a computer program product including instructions is also provided. When the computer program product is executed on a computer, the computer executes the method for collecting evidence of a vehicle changing lanes along a solid line in any of the above embodiments.

[0176] 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.

[0177] 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.

[0178] 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 system and device embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0179] 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 collecting evidence of a vehicle changing lanes on a solid line, characterized in that: The method is applied to a data processing end, wherein the data processing end communicates with a first camera, a fisheye camera, and a second camera, wherein the field of view of the first camera is the direction of the vehicle coming from each solid-line road, the field of view of the fisheye camera is the direction of looking down at each solid-line road, the field of view of the second camera is the direction of the vehicle going from each solid-line road, and the fisheye camera has overlapping acquisition areas with the first camera and the second camera respectively; the method comprises: Receive each captured information reported by each camera of the first camera and the second camera; wherein each captured information reported by each camera includes: a vehicle image captured by the camera when the camera recognizes that a vehicle enters an overlapping acquisition area with the fisheye camera, and a first type ID and coordinates of a position frame of the vehicle in the vehicle image; A first virtual queue and a second virtual queue corresponding to each frame image are obtained; wherein each frame image is collected by the fisheye camera in the process of tracking vehicles entering the field of view, and each frame image in each frame image corresponds to a vehicle queue including a second type ID and coordinates of a position frame of each vehicle in the image; the first virtual queue corresponding to each frame image is a vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into coordinates in the image coordinate system of the first camera; the second virtual queue corresponding to each frame image is a vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into coordinates in the image coordinate system of the second camera; each first type ID is an ID of a position frame of a vehicle in an image collected by any camera of the first camera and the second camera, each second type ID is an ID of a position frame of a vehicle in an image collected by the fisheye camera, and the ID of any position frame is set by the camera used to collect the image to which the position frame belongs according to the corresponding ID setting method; For each received snapshot information, according to the coordinates of the location frame in the snapshot information and the coordinates of each location frame in the virtual queue corresponding to the camera reporting the snapshot information, the first category ID of the location frame in the snapshot information is associated with the second category ID belonging to the same vehicle; wherein the virtual queue corresponding to the first camera is the first virtual queue, and the virtual queue corresponding to the second camera is the second virtual queue; In response to receiving an illegal image of any vehicle reported by the fisheye camera, a second-category ID of a position box of the vehicle in the illegal image is determined as a target ID, a first-category ID associated with the target ID is searched, and based on the illegal image and the vehicle image in the captured information to which the found first-category ID belongs, a forensic result of the vehicle's lane changing behavior on a solid line is generated; wherein the illegal image is an image captured by the fisheye camera when it is identified that any vehicle has committed an illegal lane changing behavior.

2. The method according to claim 1, characterized in that Each captured information reported by each camera also includes: the acquisition time of the vehicle image; each first virtual queue and second virtual queue are provided with the corresponding image acquisition time; The method of associating, for each received snapshot information, the first category ID of the location frame in the snapshot information with the second category ID belonging to the same vehicle according to the coordinates of the location frame in the snapshot information and the coordinates of each location frame in the virtual queue corresponding to the camera that reports the snapshot information, includes: For each received snapshot information, a virtual queue whose set acquisition time matches the acquisition time in the snapshot information is determined from the virtual queues corresponding to the camera that reported the snapshot information, as the target virtual queue; According to the coordinates of the location box in the captured information and the coordinates of each location box in the target virtual queue, the first category ID of the location box in the captured information is associated with the second category ID belonging to the same vehicle in accordance with a predetermined analysis method; wherein the predetermined analysis method is an analysis method that uses the location box overlap ratio as an analysis condition.

3. The method according to claim 2, characterized in that The step of associating the first type ID of the location frame in the captured information with the second type ID belonging to the same vehicle according to the coordinates of the location frame in the captured information and the coordinates of each location frame in the target virtual queue in a predetermined analysis manner includes: Calculating the overlap ratio of the location frame in the captured information and the location frames in the target virtual queue according to the coordinates of the location frame in the captured information and the coordinates of the location frames in the target virtual queue to obtain the overlap ratio corresponding to the location frames in the target virtual queue; The second category ID of the location box in the target virtual queue whose corresponding overlap ratio is higher than the threshold is associated with the first category ID of the location box in the snapshot information.

4. The method according to claim 2, characterized in that: For each received snapshot information, determining, from the virtual queues corresponding to the cameras that report the snapshot information, a virtual queue whose set acquisition time matches the acquisition time in the snapshot information as the target virtual queue, includes: For each snapshot information received, a virtual queue whose difference between a set acquisition time and the acquisition time in the snapshot information is within a predetermined time range is determined from the virtual queues corresponding to the camera reporting the snapshot information as the target virtual queue.

5. The method according to any one of claims 1 to 4, characterized in that: The illegal image of any vehicle includes: the image of the vehicle before changing lanes with a solid line, the image of the vehicle when changing lanes with a solid line, and the image of the vehicle after changing lanes with a solid line; The fisheye camera collects illegal images of any vehicle in the following ways: After recognizing that the vehicle has entered a solid-line road, an image of the vehicle is collected to obtain an image of the vehicle before the solid-line lane change; When the vehicle is identified as being located on a lane line of a solid line road, an image of the vehicle is captured to obtain an image of the vehicle when changing lanes along a solid line; When it is identified that the lane number of the solid lane where the vehicle is located is different from the lane number when entering the solid road, an image of the vehicle is captured to obtain an image of the vehicle after the solid line lane change.

6. The method according to any one of claims 1 to 4, characterized in that: Before generating the evidence result of the vehicle's lane-changing behavior on a solid line based on the illegal image and the vehicle image in the captured information to which the first category ID belongs, the method further includes: Acquire a target video clip captured by the fisheye camera; wherein the target video clip is a video clip whose capture time range includes the capture time of the illegal image and has a predetermined duration; The generating of the evidence collection result of the vehicle's lane changing behavior on a solid line based on the illegal image and the vehicle image in the captured information to which the first category ID belongs includes: Determine the vehicle information of the vehicle based on the vehicle image in the captured information to which the determined first category ID belongs; wherein the vehicle information of the vehicle is used to identify the vehicle; The vehicle information of the vehicle, the vehicle image in the captured information to which the first category ID is found, the illegal image and the target video clip are associated to obtain the evidence collection result of the vehicle's lane changing behavior on the solid line.

7. The method according to any one of claims 1 to 4, characterized in that: The generating of the evidence collection result of the vehicle's lane changing behavior on a solid line based on the illegal image and the vehicle image in the captured information to which the first category ID belongs includes: Based on the vehicle image in the captured information to which the first category ID belongs, a close-up image of the vehicle in the oncoming direction and a close-up image of the vehicle in the outgoing direction are generated; Determine the vehicle information of the vehicle based on the vehicle image in the captured information to which the determined first category ID belongs; wherein the vehicle information of the vehicle is used to identify the vehicle; The vehicle image in the captured information belonging to the first category ID found, the illegal image, the close-up image of the vehicle in the oncoming direction, and the close-up image of the vehicle in the outgoing direction are synthesized to obtain a synthesized image, and the obtained synthesized image is associated with the determined vehicle information to obtain the evidence collection result of the vehicle's lane changing behavior on the solid line.

8. The method according to any one of claims 1 to 4, characterized in that: The step of obtaining the first virtual queue and the second virtual queue corresponding to each frame image includes: Obtaining a vehicle queue corresponding to each frame of images reported by the fisheye camera, and for the vehicle queue corresponding to each frame of images, converting the coordinates in the vehicle queue corresponding to the frame of images into coordinates in the image coordinate system of the first camera to obtain a first virtual queue corresponding to the frame of images, and converting the coordinates in the vehicle queue corresponding to the frame of images into coordinates in the image coordinate system of the second camera to obtain a second virtual queue corresponding to the frame of images; or, Receive the first virtual queue and the second virtual queue corresponding to each frame image sent by the fisheye camera.

9. A vehicle lane-changing behavior evidence collection system, characterized in that: include: A data processing end, a first camera, a second camera and a fisheye camera; the data processing end communicates with the first camera, the fisheye camera and the second camera, the field of view of the first camera is the direction of the vehicle coming from each solid-line road, the field of view of the fisheye camera is the direction of looking down at each solid-line road, the field of view of the second camera is the direction of the vehicle going out of each solid-line road, and the fisheye camera has overlapping acquisition areas with the first camera and the second camera respectively; Each of the first camera and the second camera is used to capture the vehicle when recognizing that a vehicle enters the overlapping collection area with the fisheye camera, and report the captured information obtained by capturing the vehicle to the data processing end; wherein the captured information includes a captured vehicle image, a first type ID and coordinates of a position frame of the vehicle in the vehicle image; The fisheye camera is used to track vehicles entering the field of view, and perform image analysis and processing on each frame of images collected during the tracking process to obtain a vehicle queue of the second type ID and coordinates of a position frame of each vehicle in the image corresponding to each frame of images; The data processing end is used to receive each snapshot information reported by each camera of the first camera and the second camera, and obtain the first virtual queue and the second virtual queue corresponding to each frame image; and, for each snapshot information received, according to the coordinates of the position frame in the snapshot information and the coordinates of each position frame in the virtual queue corresponding to the camera reporting the snapshot information, associate the first type ID of the position frame in the snapshot information with the second type ID belonging to the same vehicle; wherein the virtual queue corresponding to the first camera is the first virtual queue, and the virtual queue corresponding to the second camera is the second virtual queue; the first virtual queue corresponding to each frame image is the vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into the coordinates in the image coordinate system of the first camera; the second virtual queue corresponding to each frame image is the vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into the coordinates in the image coordinate system of the second camera; each first type ID is the ID of the position frame of a vehicle in the image captured by any camera of the first camera and the second camera, and each second type ID is the ID of the position frame of a vehicle in the image captured by the fisheye camera, and the ID of any position frame is set by the camera used to capture the image to which the position frame belongs according to the corresponding ID setting method; The fisheye camera is further used to collect images when identifying any vehicle that has made an illegal lane change, obtain an illegal image of the vehicle, and report the illegal image to the data processing end; The data processing end is also used to receive the violation image, determine the second category ID of the position box of the vehicle in the violation image as the target ID, search for the first category ID associated with the target ID, and generate the evidence results of the vehicle's lane changing behavior on the solid line based on the violation image and the vehicle image in the captured information to which the found first category ID belongs.

10. A device for collecting evidence of a vehicle changing lanes on a solid line, characterized in that: Applied to a data processing end, the data processing end communicates with a first camera, a fisheye camera and a second camera, the field of view of the first camera is the direction of the vehicle coming from each solid-line road, the field of view of the fisheye camera is the direction of looking down at each solid-line road, the field of view of the second camera is the direction of the vehicle going from each solid-line road, and the fisheye camera has overlapping acquisition areas with the first camera and the second camera respectively; the device comprises: A first receiving module is used to receive each captured information reported by each camera of the first camera and the second camera; wherein each captured information reported by each camera includes: a vehicle image captured by the camera when the camera recognizes that a vehicle enters the overlapping acquisition area with the fisheye camera, and a first type ID and coordinates of a position frame of the vehicle in the vehicle image; an acquisition module, for acquiring a first virtual queue and a second virtual queue corresponding to each frame image; wherein each frame image is collected by the fisheye camera in the process of tracking vehicles entering the field of view, and each frame image in each frame image corresponds to a vehicle queue including a second type ID and coordinates of a position frame of each vehicle in the image; the first virtual queue corresponding to each frame image is a vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into coordinates in the image coordinate system of the first camera; the second virtual queue corresponding to each frame image is a vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into coordinates in the image coordinate system of the second camera; each first type ID is an ID of a position frame of a vehicle in an image collected by any camera of the first camera and the second camera, each second type ID is an ID of a position frame of a vehicle in an image collected by the fisheye camera, and the ID of any position frame is set by the camera used to collect the image to which the position frame belongs according to the corresponding ID setting method; an associating module, for associating, for each received snapshot information, a first category ID of the location frame in the snapshot information with a second category ID belonging to the same vehicle according to the coordinates of the location frame in the snapshot information and the coordinates of each location frame in the virtual queue corresponding to the camera reporting the snapshot information; wherein the virtual queue corresponding to the first camera is the first virtual queue, and the virtual queue corresponding to the second camera is the second virtual queue; A generation module is used to, in response to receiving an illegal image of any vehicle reported by the fisheye camera, determine the second category ID of the position box of the vehicle in the illegal image as the target ID, search for the first category ID associated with the target ID, and generate a forensic result of the vehicle's lane changing behavior on a solid line based on the illegal image and the vehicle image in the captured information to which the found first category ID belongs; wherein the illegal image is an image captured by the fisheye camera when it is identified that any vehicle has made an illegal lane changing behavior.

11. A data processing device, characterized in that: The data processing device communicates with a first camera, a fisheye camera and a second camera, the field of view of the first camera is the direction of the vehicle coming from each solid-line road, the field of view of the fisheye camera is the direction of looking down at each solid-line road, the field of view of the second camera is the direction of the vehicle going out of each solid-line road, and the fisheye camera has overlapping acquisition areas with the first camera and the second camera respectively; the data processing device comprises: Memory, used to store computer programs; A processor, for implementing any of the methods described in claims 1-8 when executing a program stored in a memory.

12. An all-in-one device, characterized in that: The system comprises a first camera, a fisheye camera, a second camera and a data processing end; the data processing end communicates with the first camera, the fisheye camera and the second camera, the field of view of the first camera is the direction of the vehicle coming from each solid-line road, the field of view of the fisheye camera is the direction of looking down at each solid-line road, the field of view of the second camera is the direction of the vehicle going out of each solid-line road, and the fisheye camera has overlapping acquisition areas with the first camera and the second camera respectively; Each of the first camera and the second camera is used to capture the vehicle when recognizing that a vehicle enters the overlapping collection area with the fisheye camera, and report the captured information obtained by capturing the vehicle to the data processing end; wherein the captured information includes a captured vehicle image, a first type ID and coordinates of a position frame of the vehicle in the vehicle image; The fisheye camera is used to track vehicles entering the field of view, and perform image analysis and processing on each frame of images collected during the tracking process to obtain a vehicle queue of the second type ID and coordinates of a position frame of each vehicle in the image corresponding to each frame of images; The data processing end is used to receive each snapshot information reported by each camera of the first camera and the second camera, and obtain the first virtual queue and the second virtual queue corresponding to each frame image; and, for each snapshot information received, according to the coordinates of the position frame in the snapshot information and the coordinates of each position frame in the virtual queue corresponding to the camera reporting the snapshot information, associate the first type ID of the position frame in the snapshot information with the second type ID belonging to the same vehicle; wherein the virtual queue corresponding to the first camera is the first virtual queue, and the virtual queue corresponding to the second camera is the second virtual queue; the first virtual queue corresponding to each frame image is the vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into the coordinates in the image coordinate system of the first camera; the second virtual queue corresponding to each frame image is the vehicle queue formed after the coordinates in the vehicle queue corresponding to the image are converted into the coordinates in the image coordinate system of the second camera; each first type ID is the ID of the position frame of a vehicle in the image captured by any camera of the first camera and the second camera, and each second type ID is the ID of the position frame of a vehicle in the image captured by the fisheye camera, and the ID of any position frame is set by the camera used to capture the image to which the position frame belongs according to the corresponding ID setting method; The fisheye camera is further used to collect images when identifying any vehicle that has made an illegal lane change, obtain an illegal image of the vehicle, and report the illegal image to the data processing end; The data processing end is also used to receive the violation image, determine the second category ID of the position box of the vehicle in the violation image as the target ID, search for the first category ID associated with the target ID, and generate the evidence results of the vehicle's lane changing behavior on the solid line based on the violation image and the vehicle image in the captured information to which the found first category ID belongs.

13. 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 8 is implemented.

Images