A vehicle driving state detection method and device based on monocular camera

Abstract

The present invention is applicable to the field of intelligent transportation technology, and provides a vehicle driving state detection method and device based on a monocular camera. In this embodiment, after the driving monitoring information of the target vehicle is obtained by a monocular camera, the initial position information and the target position information of the target vehicle are determined by two video frames in the information to determine the moving distance of the target vehicle between the two video frames; then the driving parameters of the target vehicle are determined according to the moving distance, the upper and lower offsets T of the monocular camera, the magnification Z, the vertical height H and other parameters, and finally the driving state of the vehicle is determined by the driving parameters. This solves the problems existing in the existing vehicle driving state monitoring method, such as the error caused by the viewing angle, the high license plate error at a distance, and the need for multiple dense intrusion lines to alleviate the problem due to the camera angle. It is realized that the position correlation relationship of the vehicle in the picture is judged by the monocular camera, so as to judge the driving state of the vehicle.

Description

Technical Field

[0001] The present application relates to the field of intelligent transportation technology, and in particular to a vehicle driving status detection method and device based on a monocular camera. Background Art

[0002] In the context of intelligent transportation systems (ITS), the detection of vehicle driving status is one of the key technologies to ensure traffic safety and efficiency. With the acceleration of urbanization and the sharp increase in the number of cars, urban traffic management faces unprecedented challenges. Therefore, how to effectively monitor the driving status of vehicles has become a key issue in intelligent transportation systems.

[0003] At present, the existing vehicle driving status monitoring methods are basically divided into the following three types:

[0004] (I) Stereo geometry and motion assumptions: The homography method of projecting 3D space into 2D space involves the calibration of the camera's internal parameters (including focal length f and distortion matrix, etc.), and the connection points between the vehicle and the road surface.

[0005] (ii) Based on prior knowledge of real object size. It can be roughly divided into two categories: known vehicle size and known license plate size. Distance estimation based on license plate size is very sensitive to meter-pixel ratio and target positioning, and requires retraining the model for recognizing license plates. Distance estimation based on vehicle size requires relying on a fine-grained vehicle model classification system to determine the specific vehicle model and use its accurate size.

[0006] (iii) Intrusion lines or areas. This approach does not require calibration of the camera system, but requires the actual distance between two or more virtual lines on the road, or the actual size of the area. Then, the distance estimation problem is converted to detecting which vehicles pass through the pre-defined lines or areas.

[0007] However, all of the above methods have significant technical drawbacks.

[0008] For method (I) of stereo geometry and motion assumptions, this method will lead to certain errors due to the system perspective problem. Figure 1 As shown, Figure 1 The viewing angle in (a) will cause errors. It is the actual value. The connection point is not visible at t+N, which leads to a certain error in the estimated value. Figure 1 In (b), judging the distance moved by the license plate will lead to an incorrect judgment of the vehicle's moving distance. is the actual value, and is the incorrect estimated distance after using homography conversion.

[0009] As for method (2), it is based on the prior knowledge of the real object size. This method uses the license plate size to estimate the distance, which is very sensitive to the meter-pixel ratio and target positioning. That is, the error is high when the license plate is not recognizable at a distance, and the model for recognizing the license plate needs to be retrained. Estimating the distance by vehicle size requires relying on a fine-grained vehicle classification system to determine the specific vehicle model and use its accurate size.

[0010] For method (iii), intrusion lines or areas, this method is limited by camera angle, space (camera resolution and distance between camera and vehicle), time (camera frame rate and vehicle speed), etc., and requires multiple dense intrusion lines to mitigate. Summary of the invention

[0011] In view of this, the present application provides a vehicle driving status detection method and device based on a monocular camera, so as to determine the position correlation relationship of the vehicle in the picture through the monocular camera, thereby determining the driving status of the vehicle, thereby completing the practical application of monocular camera vehicle speed measurement.

[0012] The first aspect of the present application provides a vehicle driving state detection method based on a monocular camera, which is applied to an intelligent transportation system. The method comprises:

[0013] Acquire driving monitoring information of the target vehicle through the monocular camera, wherein the driving monitoring information at least includes a first video frame and a second video frame shot later than the first video frame;

[0014] Determine the initial position information of the target vehicle through the first video frame, and determine the target position information of the target vehicle after traveling through the second video frame;

[0015] Determine the moving distance of the target vehicle between the first video frame and the second video frame by using the initial position information and the target position information;

[0016] Determine the parameter information of the monocular camera, and determine the driving parameter β of the target vehicle through the target position information, the moving distance and the parameter information, and then determine the driving state of the target vehicle through the driving parameter β, wherein the parameter information includes the up and down offset T, the magnification Z, and the vertical height H.

[0017] Optionally, the parameter information further includes a left-right offset P, and the method further includes:

[0018] The lane direction of the target vehicle is determined by the monocular camera, and whether the target vehicle is traveling in the reverse direction is determined by the initial position information, the target position information, the left and right offsets P and the lane direction, and an alarm message is sent to the system administrator when it is determined.

[0019] Optionally, determining the initial position information of the target vehicle through the first video frame, and determining the target position information of the target vehicle after traveling through the second video frame includes:

[0020] Construct a world coordinate system, determine a first contour image of the target vehicle in the first video frame, and determine the maximum value x of the X coordinate of the first contour image in the world coordinate system. 1max , minimum value x 1min And the maximum value of the Y coordinate y 1max , minimum value y 1min , and [x 1min ,y 1min , x 1max ,y 1max ] is determined as the initial position information of the target vehicle;

[0021] Determine a second contour image of the target vehicle in the second video frame, and determine a maximum value x of the X coordinate of the second contour image in the world coordinate system. 2max , minimum value x 2min And the maximum value of the Y coordinate y 2max , minimum value y 2min , and [x 2min ,y 2min , x 2max ,y 2max ] is determined as the target position information of the target vehicle.

[0022] Optionally, determining the driving parameter β of the target vehicle by using the target position information, the moving distance and the parameter information includes:

[0023] The center point coordinates (x 2mid ,y 2mid ), and then through the formula The driving parameter β is determined, wherein distance is the moving distance.

[0024] Optionally, determining the driving state of the target vehicle by using the driving parameter β includes:

[0025] Pre-set the distance threshold threshold1 and distance threshold threshold2, through the formula Determine β1 corresponding to the distance threshold threshold1 and β2 corresponding to the distance threshold threshold2, wherein the y 2mid is the Y center coordinate of the target vehicle in the second video frame;

[0026] The driving parameter β is compared with the β1 and β2, and the driving state of the target vehicle is determined according to the comparison result.

[0027] The second aspect of the present application provides a vehicle driving state detection device based on a monocular camera, which is applied to an intelligent transportation system, and the device includes:

[0028] An information acquisition unit, configured to acquire driving monitoring information of a target vehicle through the monocular camera, wherein the driving monitoring information at least includes a first video frame and a second video frame shot later than the first video frame;

[0029] a position information determining unit, configured to determine the initial position information of the target vehicle through the first video frame, and determine the target position information of the target vehicle after the vehicle has traveled through the second video frame;

[0030] a moving distance determining unit, configured to determine a moving distance of the target vehicle between the first video frame and the second video frame according to the initial position information and the target position information;

[0031] A driving state determination unit is used to determine the parameter information of the monocular camera, and determine the driving parameter β of the target vehicle through the target position information, the moving distance and the parameter information, and then determine the driving state of the target vehicle through the driving parameter β, wherein the parameter information includes the up and down offset T, the magnification Z, and the vertical height H.

[0032] Optionally, the parameter information in the driving state determination unit further includes left and right offset P, and the device further includes:

[0033] A reverse traffic determination unit is used to determine the lane direction of the target vehicle through the monocular camera, determine whether the target vehicle is reverse traffic through the initial position information, target position information, left and right offset P and the lane direction, and send an alarm message to the system administrator when it is determined.

[0034] Optionally, the determining of the initial position information of the target vehicle by the first video frame and the determining of the target position information after the target vehicle travels by the second video frame in the position information determining unit includes:

[0035] Construct a world coordinate system, determine a first contour image of the target vehicle in the first video frame, and determine the maximum value x of the X coordinate of the first contour image in the world coordinate system. 1max , minimum value x 1min And the maximum value of the Y coordinate y 1max , minimum value y 1min , and [x 1min ,y 1min, x 1max ,y 1max ] is determined as the initial position information of the target vehicle;

[0036] Determine a second contour image of the target vehicle in the second video frame, and determine a maximum value x of the X coordinate of the second contour image in the world coordinate system. 2max , minimum value x 2min And the maximum value of the Y coordinate y 2max , minimum value y 2min , and [x 2min ,y 2min , x 2max ,y 2max ] is determined as the target position information of the target vehicle.

[0037] Optionally, the driving state determining unit determines the driving parameter β of the target vehicle by using the target position information, the moving distance and the parameter information, including:

[0038] The center point coordinates (x 2mid ,y 2mid ), and then through the formula The driving parameter β is determined, wherein distance is the moving distance.

[0039] Optionally, the driving state determining unit determines the driving state of the target vehicle by using the driving parameter β, including:

[0040] Pre-set the distance threshold threshold1 and distance threshold threshold2, through the formula Determine β1 corresponding to the distance threshold threshold1 and β2 corresponding to the distance threshold threshold2, wherein the y 2mid is the Y center coordinate of the target vehicle in the second video frame;

[0041] The driving parameter β is compared with the β1 and β2, and the driving state of the target vehicle is determined according to the comparison result.

[0042] In the embodiment provided in the present application, after the driving monitoring information of the target vehicle is obtained through a monocular camera, the initial position information and the target position information of the target vehicle are determined through the two video frames in the information to determine the moving distance of the target vehicle between the two video frames; then the driving parameters of the target vehicle are determined according to the moving distance, the upper and lower offsets T of the monocular camera, the magnification Z, the vertical height H and other parameters, and finally the driving state of the vehicle is determined through the driving parameters. This solves the problems existing in the existing vehicle driving state monitoring method, such as the error caused by the viewing angle, the high license plate error at a distance, and the need for multiple dense intrusion lines to alleviate the problem due to the limitation of the camera angle. It is realized that the position correlation relationship of the vehicle in the picture is judged by the monocular camera, so as to judge the driving state of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a spatial projection diagram of the solid geometry and motion hypothesis method in the background technology;

[0044] Figure 2 A flow chart of the method provided in the embodiment of the present application;

[0045] Figure 3 A contour coordinate diagram provided for an embodiment of the present application;

[0046] Figure 4 A structural diagram of a device provided in an embodiment of the present application;

[0047] Figure 5 A schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0049] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0050] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0051] The present application provides a vehicle driving state detection method and device based on a monocular camera, so as to determine the position correlation relationship of the vehicle in the picture through the monocular camera, thereby determining the driving state of the vehicle.

[0052] The technical solution of the present application is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0053] like Figure 2 As shown, it is a flow chart of a vehicle driving state detection method based on a monocular camera provided by the present application, and the process may include the following steps:

[0054] Step S201: Acquire driving monitoring information of a target vehicle through the monocular camera, wherein the driving monitoring information at least includes a first video frame and a second video frame shot later than the first video frame.

[0055] In this embodiment, the driving monitoring information can be a video stream of the target vehicle captured by the monocular camera, and the video stream is composed of continuous video frames, each of which is a static image. During the capture process, the system will give each video a timestamp to ensure that the shooting time of each frame can be tracked. In this way, when a video frame in the video stream is determined to be the first video frame, the second video frame that is later than the first video frame can be determined from the video stream through the timestamp.

[0056] In this embodiment, two consecutive video frames may also be selected as the first video frame and the second video frame, that is, after the first video frame (assuming it is frame A) is extracted, the next frame is naturally the second video frame.

[0057] Step S202: determining the initial position information of the target vehicle through the first video frame, and determining the target position information of the target vehicle after driving through the second video frame.

[0058] In this embodiment, a vehicle can be detected in a video frame by a computer vision algorithm. First, the vehicle is identified on the video frame and feature points are extracted, such as the four corners of the vehicle, the license plate position, or other significant features. Then, the vehicle is identified based on a convolutional neural network (CNN) to determine the initial position information and target position information of the vehicle through the first and second video frames.

[0059] In another embodiment, determining the initial position information of the target vehicle through the first video frame, and determining the target position information of the target vehicle after traveling through the second video frame includes:

[0060] Construct a world coordinate system, determine a first contour image of the target vehicle in the first video frame, and determine the maximum value x of the X coordinate of the first contour image in the world coordinate system. 1max , minimum value x 1min And the maximum value of the Y coordinate y 1max , minimum value y 1min , and [x 1min ,y 1min , x 1max ,y 1max ] is determined as the initial position information of the target vehicle;

[0061] Determine a second contour image of the target vehicle in the second video frame, and determine a maximum value x of the X coordinate of the second contour image in the world coordinate system. 2max , minimum value x 2min And the maximum value of the Y coordinate y 2max , minimum value y 2min , and [x 2min ,y 2min , x 2max ,y 2max ] is determined as the target position information of the target vehicle.

[0062] In this embodiment, a coordinate system of the monitoring area is pre-constructed, and then edge detection and contour extraction operations are performed on the first and second video frames to determine the first and second contour images of the target vehicle. By combining the contour image with the coordinate system, the maximum and minimum values ​​of the X and Y coordinates on the contour image can be determined, such as Figure 3 As shown in the figure, (x1, y1) is the first outline of the target vehicle in the first video frame, and (x2, y2) is the second outline of the target vehicle in the second video frame. Since the maximum and minimum values ​​of the X and Y coordinates represent the boundary points of the target vehicle, they can be determined as the location information of the target vehicle. That is, [x 1min ,y 1min , x 1max ,y 1max ] is the initial position information of the target vehicle; [x2min ,y 2min , x 2max ,y 2max ] is the target position information of the target vehicle.

[0063] Step S203: determining a moving distance of the target vehicle between the first video frame and the second video frame according to the initial position information and the target position information.

[0064] In this embodiment, the center point position of the target vehicle in the first and second video frames can be determined by the initial position information and the target position information, and then the two center point positions are placed on the same plane to calculate the moving distance between the first and second video frames.

[0065] For example, the initial position information is [x 1min ,y 1min , x 1max ,y 1max ], the target location information is [x 2min ,y 2min , x 2max ,y 2max ] as an example, the center point of the two positions can be determined as (x 2mid ,y 2mid ), then the distance between the first and second video frames can be determined by the formula, and this distance is the moving distance of the target vehicle.

[0066] Step S204, determining the parameter information of the monocular camera, and determining the driving parameter β of the target vehicle through the target position information, the moving distance and the parameter information, and then determining the driving state of the target vehicle through the driving parameter β.

[0067] In this embodiment, the parameter information of the monocular camera includes the up-down offset T, the magnification Z, and the vertical height H. The up-down offset T is the pitch angle of the monocular camera, which is a fixed value range. An initial value can be determined in advance at the installation angle of the monocular camera. When the angle of the monocular camera is greater than the initial angle, the up-down offset T increases, and vice versa. The magnification Z and the vertical height H can be directly obtained by reading the camera parameters and measuring the height.

[0068] In this embodiment, several different β thresholds can be pre-set based on historical data, and then the center point coordinates (x 2mid ,y 2mid ), and then according to the formula Determine the driving parameter β. By comparing β with the β threshold, the judgment of stopping, slowing down, going against the flow and speeding can be realized.

[0069] In another embodiment, determining the driving state of the target vehicle by using the driving parameter β includes:

[0070] Pre-set the distance threshold threshold1 and distance threshold threshold2, through the formula Determine β1 corresponding to the distance threshold threshold1 and β2 corresponding to the distance threshold threshold2, wherein the Y 2mid is the Y center coordinate of the target vehicle in the second video frame;

[0071] The driving parameter β is compared with β1 and β2, and the driving state of the target vehicle is determined according to the comparison result.

[0072] In this embodiment, the distance threshold threshold1 can be the minimum distance threshold determined based on historical data, and the distance threshold threshold2 is the distance threshold for normal driving judgment. On this basis, β1 determined by the distance threshold threshold1 can be regarded as the minimum driving speed of the target vehicle, and β2 is the normal driving speed. When the driving parameter β is less than β1, the driving state of the target vehicle is determined to be a stopped state; when β is between β1 and β2, the driving state of the target vehicle is determined to be a slow-moving state; when β is greater than β2, the driving state of the target vehicle is determined to be a normal state.

[0073] Furthermore, in this embodiment, a distance threshold value threshold3 for speeding judgment may be preset, and a parameter β3 for speeding judgment may be determined by the above formula. When β is greater than β3, the driving state of the target vehicle is determined to be a speeding state.

[0074] In another embodiment, the parameter information further includes a left-right offset P, and the method further includes:

[0075] The lane direction of the target vehicle is determined by the monocular camera, and whether the target vehicle is traveling in the reverse direction is determined by the initial position information, the target position information, the left and right offsets P and the lane direction, and an alarm message is sent to the system administrator when it is determined.

[0076] In this embodiment, the left-right offset P is used to characterize the left-right moving distance of the monocular camera from the installation position, which is a fixed range of values. An initial value can be determined in advance according to the installation position of the monocular camera. When the monocular camera moves to the left, the left-right offset P increases, and vice versa. It can be determined whether the target vehicle is going in the opposite direction by the lane direction and the positive and negative results of calculating (P-180)*(y1mid-y2mid). For example, when it is judged by the lane direction of the target vehicle that it is traveling to the left, and the above calculation result is positive, it can be determined that the vehicle is not going in the opposite direction. When it is judged that it is traveling to the left and the above calculation result is positive, it can be determined that the target vehicle is going in the opposite direction.

[0077] So far, completed Figure 2 The process shown.

[0078] In the embodiment of the present application, after the driving monitoring information of the target vehicle is obtained through a monocular camera, the initial position information and the target position information of the target vehicle are determined through the two video frames in the information to determine the moving distance of the target vehicle between the two video frames; then the driving parameters of the target vehicle are determined according to the moving distance, the upper and lower offset T of the monocular camera, the magnification Z, the vertical height H and other parameters, and finally the driving state of the vehicle is determined through the driving parameters. This solves the problems existing in the existing vehicle driving state monitoring method, such as the error caused by the viewing angle, the high error of the license plate at a distance, and the need for multiple dense intrusion lines to alleviate the problem due to the limitation of the camera angle. It is realized that the position correlation relationship of the vehicle in the picture is judged by the monocular camera, so as to judge the driving state of the vehicle.

[0079] like Figure 4 As shown, the present application also provides a vehicle driving state detection device based on a monocular camera, which is applied to an intelligent transportation system, and the device includes:

[0080] An information acquisition unit 401 is used to acquire driving monitoring information of a target vehicle through the monocular camera, wherein the driving monitoring information at least includes a first video frame and a second video frame shot later than the first video frame;

[0081] A position information determining unit 402, configured to determine the initial position information of the target vehicle through the first video frame, and determine the target position information of the target vehicle after the vehicle has traveled through the second video frame;

[0082] A moving distance determining unit 403, configured to determine a moving distance of the target vehicle between the first video frame and the second video frame according to the initial position information and the target position information;

[0083] The driving state determination unit 404 is used to determine the parameter information of the monocular camera, and determine the driving parameter β of the target vehicle through the target position information, the moving distance and the parameter information, and then determine the driving state of the target vehicle through the driving parameter β, wherein the parameter information includes the upper and lower offsets T, the magnification Z, and the vertical height H.

[0084] In another embodiment, the parameter information in the driving state determination unit further includes a left-right offset P, and the device further includes:

[0085] The reverse traffic determination unit 405 is used to determine the lane direction of the target vehicle through the monocular camera, determine whether the target vehicle is reverse traffic through the initial position information, target position information, left and right offset P and the lane direction, and send an alarm message to the system administrator when it is determined.

[0086] In another embodiment, the determining unit determines the initial position information of the target vehicle through the first video frame, and determines the target position information of the target vehicle after traveling through the second video frame, including:

[0087] Construct a world coordinate system, determine a first contour image of the target vehicle in the first video frame, and determine the maximum value x of the X coordinate of the first contour image in the world coordinate system. 1max , minimum value x 1min And the maximum value of the Y coordinate y 1max , minimum value y 1min , and [x 1min ,y 1min , x 1max ,y 1max ] is determined as the initial position information of the target vehicle;

[0088] Determine a second contour image of the target vehicle in the second video frame, and determine a maximum value x of the X coordinate of the second contour image in the world coordinate system. 2max , minimum value x 2min And the maximum value of the Y coordinate y 2max , minimum value y 2min , and [x 2min ,y 2min , x 2max ,y 2max ] is determined as the target position information of the target vehicle.

[0089] In another embodiment, the driving state determining unit determines the driving parameter β of the target vehicle by using the target position information, the moving distance and the parameter information, including:

[0090] The center point coordinates (x 2mid ,y 2mid ), and then through the formula The driving parameter β is determined, wherein distance is the moving distance.

[0091] In another embodiment, determining the driving state of the target vehicle by using the driving parameter β in the driving state determination unit includes:

[0092] Pre-set the distance threshold threshold1 and distance threshold threshold2, through the formula Determine β1 corresponding to the distance threshold threshold1 and β2 corresponding to the distance threshold threshold2, wherein the y 2mid is the Y center coordinate of the target vehicle in the second video frame;

[0093] The driving parameter β is compared with the β1 and β2, and the driving state of the target vehicle is determined according to the comparison result.

[0094] The above-mentioned embodiment of the present invention provides a vehicle driving state detection method based on a monocular camera, and based on this method, provides a vehicle driving state detection device based on a monocular camera. Through the above-mentioned method and device, it is possible to judge the position association relationship of the vehicle in the picture through the monocular camera, thereby judging the driving state of the vehicle.

[0095] This embodiment also discloses a computer device, such as Figure 5 As shown, the computer device includes a processor and a memory, wherein the memory stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement any of the above-mentioned vehicle driving state detection methods based on a monocular camera.

[0096] In addition, in the above-mentioned example of the vehicle driving state detection device based on a monocular camera, the logical division of each program module is only an example. In actual application, the above-mentioned functions can be assigned to different program modules as needed, for example, for the convenience of corresponding hardware configuration requirements or software implementation. That is, the internal structure of the vehicle driving state detection device based on a monocular camera is divided into different program modules to complete all or part of the functions described above.

[0097] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A vehicle driving state detection method based on a monocular camera, characterized in that: Applied to an intelligent transportation system, the method comprises: Acquire driving monitoring information of the target vehicle through the monocular camera, wherein the driving monitoring information at least includes a first video frame and a second video frame shot later than the first video frame; Construct a world coordinate system, determine a first contour image of the target vehicle in the first video frame, and determine the maximum value x of the X coordinate of the first contour image in the world coordinate system. 1max , minimum value x 1min And the maximum value of the Y coordinate y 1max , minimum value y 1min , and [x 1min ,y 1min , x 1max ,y 1max ] is determined as the initial position information of the target vehicle; Determine a second contour image of the target vehicle in the second video frame, and determine a maximum value x of the X coordinate of the second contour image in the world coordinate system. 2max , minimum value x 2min And the maximum value of the Y coordinate y 2max , minimum value y 2min , and [x 2min ,y 2min , x 2max ,y 2max ] is determined as the target position information of the target vehicle; Determine the moving distance of the target vehicle between the first video frame and the second video frame by using the initial position information and the target position information; Determine the parameter information of the monocular camera, and determine the center point coordinates (x 2mid ,y 2mid ), the parameter information includes up and down offset T, magnification Z, and vertical height H; Pre-set the distance threshold threshold1 and distance threshold threshold2, through the formula Determine the driving parameter β, β1 corresponding to the distance threshold threshold1, and β2 corresponding to the distance threshold threshold2, compare the driving parameter β with the β1 and β2, and when the driving parameter β is less than β1, determine that the driving state of the target vehicle is a stopped state; when the driving parameter β is between β1 and β2, determine that the driving state of the target vehicle is a slow moving state; when the driving parameter β is greater than β2, determine that the driving state of the target vehicle is a normal state, wherein β1 is a minimum driving speed, β2 is a normal driving speed, and distance is the moving distance.

2. The method according to claim 1, characterized in that The parameter information also includes a left-right offset P, and the method further includes: The lane direction of the target vehicle is determined by the monocular camera, and whether the target vehicle is traveling in the reverse direction is determined by the initial position information, the target position information, the left and right offsets P and the lane direction, and an alarm message is sent to the system administrator when it is determined.

3. A vehicle driving state detection device based on a monocular camera, characterized in that: Applied to an intelligent transportation system, the device comprises: An information acquisition unit, configured to acquire driving monitoring information of a target vehicle through the monocular camera, wherein the driving monitoring information at least includes a first video frame and a second video frame shot later than the first video frame; A position information determination unit is used to construct a world coordinate system, determine a first contour image of the target vehicle in the first video frame, and determine a maximum value x of an X coordinate of the first contour image in the world coordinate system. 1max , minimum value x 1min And the maximum value of the Y coordinate y 1max , minimum value y 1min , and [x 1min ,y 1min , x 1max ,y 1max ] is determined as the initial position information of the target vehicle; Determine a second contour image of the target vehicle in the second video frame, and determine a maximum value x of the X coordinate of the second contour image in the world coordinate system. 2max , minimum value x 2min And the maximum value of the Y coordinate y 2max , minimum value y 2min , and [x 2min ,y 2min , x 2max ,y 2max ] is determined as the target position information of the target vehicle; a moving distance determining unit, configured to determine a moving distance of the target vehicle between the first video frame and the second video frame according to the initial position information and the target position information; A driving state determination unit is used to determine the parameter information of the monocular camera, and determine the center point coordinates (x 2mid ,y 2mid ), the parameter information includes up and down offset T, magnification Z, and vertical height H; Pre-set the distance threshold threshold1 and distance threshold threshold2, through the formula Determine the driving parameter β, β1 corresponding to the distance threshold threshold1, and β2 corresponding to the distance threshold threshold2, compare the driving parameter β with the β1 and β2, and when the driving parameter β is less than β1, determine that the driving state of the target vehicle is a stopped state; when the driving parameter β is between β1 and β2, determine that the driving state of the target vehicle is a slow moving state; when the driving parameter β is greater than β2, determine that the driving state of the target vehicle is a normal state, and determine that the driving state of the target vehicle is a normal state, wherein β1 is a minimum driving speed, β2 is a normal driving speed, and distance is the moving distance.

4. The device according to claim 3, characterized in that The parameter information in the driving state determination unit also includes a left-right offset P, and the device also includes: A reverse traffic determination unit is used to determine the lane direction of the target vehicle through the monocular camera, determine whether the target vehicle is reverse traffic through the initial position information, target position information, left and right offset P and the lane direction, and send an alarm message to the system administrator when it is determined.

Images