Vehicle detection method and system, and computer readable storage medium

Through the vehicle detection model, the position and motion status of the wiper are identified, the authenticity of the vehicle video is verified, and the problems of inconsistency and cheating in the online car-hailing industry are solved, and safety guarantees are improved.

CN111860127BActive Publication Date: 2024-11-05BEIJING DIDI INFINITY TECH & DEV CO LTD
View PDF 9 Cites 0 Cited by

Patent Information

Application Number
CN202010503138.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-06-05
Publication Date
2024-11-05
Estimated Expiration
2040-06-05

AI Technical Summary

Technical Problem

In the online car-hailing industry, the vehicle reservations made by users are inconsistent with the vehicle information displayed on the platform, which leads to the threat of safety protection. Some drivers use cheating methods such as submitting old videos or remakes of screens, resulting in inaccurate verification of vehicle information.

Method used

The vehicle detection model receives the video information uploaded by the driver and identifies the position and motion state of the wiper. If the wiper is operating normally, it is considered that the video is a real vehicle video taken in real time. Otherwise, it is determined to be cheating, and the authenticity of the vehicle is verified.

Benefits of technology

Effectively distinguish whether the uploaded video is a live vehicle video taken by real-life video, which improves the accuracy of vehicle information verification, prevents cheating, and ensures the authenticity of vehicle information.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN111860127B_ABST
    Figure CN111860127B_ABST
Patent Text Reader

Abstract

The present invention provides a vehicle detection method and system, and a computer-readable storage medium, wherein the vehicle detection method includes: receiving video information; determining the vehicle image information corresponding to any frame image in the video information through a vehicle detection model, and determining the position information of the wiper of the vehicle to be detected based on the vehicle image information; and determining that the vehicle to be detected has passed the detection based on the position information satisfying a preset condition. The technical solution provided by the present invention determines whether the vehicle video uploaded by the driver is a real "live" vehicle video based on the motion state of the wiper when detecting the vehicle video uploaded by the driver, and can effectively distinguish cheating methods such as uploading old videos and reshooting the screen, thereby improving the accuracy of video detection.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle detection, and in particular to a vehicle detection method, a vehicle detection system and a computer-readable storage medium. Background Art

[0002] In the related technology, in the online car-hailing industry, there are cases where the vehicle booked by the user is inconsistent with the vehicle information displayed on the platform. In order to avoid this situation and improve safety, the platform needs to verify whether the vehicle driven by the driver is consistent with the registered vehicle information. During the verification process, some drivers cheat by submitting previously recorded videos or reshooting the screen.

[0003] Therefore, there is an urgent need for a detection method to ensure that the vehicle information submitted by the driver is authentic and valid. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art or related art.

[0005] To this end, a first aspect of the present invention provides a vehicle detection method.

[0006] A second aspect of the present invention provides a vehicle detection system.

[0007] A third aspect of the present invention provides a computer-readable storage medium.

[0008] In view of this, the first aspect of the present invention provides a vehicle detection method, including: receiving video information; determining the vehicle image information corresponding to any frame image in the video information through a vehicle detection model, and determining the position information of the wiper of the vehicle to be detected based on the vehicle image information; based on the position information satisfying preset conditions, determining that the vehicle to be detected has passed the detection.

[0009] In this technical solution, when testing a vehicle, the real video information uploaded by the driver is first received. When the driver is asked to shoot a video of the vehicle, the wipers are turned on at the same time. If the wipers are detected to be operating normally in the video uploaded by the driver, the video uploaded by the driver is considered to be a real vehicle video shot in real time. If the wipers are detected not to be operating normally in the video uploaded by the driver, it indicates that the driver is likely to cheat, and the vehicle in the uploaded video is not a real "live" vehicle, and the vehicle detection fails at this time.

[0010] Specifically, the received video information is input into the vehicle detection model, and the video information is processed by the vehicle detection model to obtain continuous frame images, and each frame image is further processed into corresponding vehicle image information, and the position information of the vehicle's windshield wiper in the vehicle image information is determined frame by frame. When the windshield wiper meets the preset conditions built into the vehicle detection model, it is determined that the windshield wiper operates normally as required, and at this time it is determined that the vehicle to be detected has passed the detection.

[0011] The technical solution provided by the present invention determines whether the vehicle video uploaded by the driver is a real "live" vehicle video according to the movement state of the wiper when detecting the vehicle video uploaded by the driver. It can effectively distinguish cheating methods such as uploading old videos and reshooting the screen, thereby improving the accuracy of video detection.

[0012] In addition, the vehicle detection method in the above technical solution provided by the present invention may also have the following additional technical features:

[0013] In the above technical solution, the wiper includes a rotating arm and a brush body, and the vehicle to be detected also includes a front windshield; the step of determining the position information of the wiper of the vehicle to be detected based on the vehicle image information specifically includes: determining the first intersection point of the side where the rotating arm intersects with the front windshield through the vehicle detection model, determining the second intersection point of the rotating arm and the brush body, and determining the first end point of the brush body; determining the position information based on the first intersection point, the second intersection point and the first end point.

[0014] In this technical solution, the wiper of a general vehicle includes a rotating arm and a brush body. After being identified by the vehicle detection model, the front windshield of the vehicle is identified as an approximately rectangular area, and the wiper moves within the rectangular area. Specifically, the point where the wiper intersects with the bottom edge of the rectangular area corresponding to the front windshield is the root node of the wiper, that is, the first intersection point, and the wiper rotates approximately around the root node. The position where the rotating arm of the wiper intersects with the brush body is the second intersection point, and the brush body rotates relative to the rotating arm around the second intersection point. At the same time, the brush body itself is processed into a shape that approximates a line segment, and its two ends are marked as the first endpoint and the second endpoint respectively.

[0015] It can be obtained that the first intersection and the second intersection are on the same straight line, that is, they are on the same arm of the wiper. The second intersection and the first endpoint are on the same straight line, that is, the brush body of the wiper. Therefore, the positions of the wiper arm and the brush body can be determined through the first intersection, the second intersection and the first endpoint, and then the position information of the wiper can be obtained.

[0016] In any of the above technical solutions, the step of determining the position information based on the first intersection, the second intersection and the first endpoint specifically includes: determining the first corner point, the second corner point, the third corner point and the fourth corner point corresponding to the windshield through a vehicle detection model; establishing a Cartesian coordinate system with any one of the first corner point, the second corner point, the third corner point and the fourth corner point as the origin, and establishing an angular coordinate system with the first intersection as the origin; determining the angle data between the side where the windshield intersects with the swing arm and the swing arm in the angular coordinate system; determining the first coordinate corresponding to the first endpoint and the second coordinate corresponding to the second intersection in the Cartesian coordinate system; determining the position information based on the angle data, the first coordinate and the second coordinate.

[0017] In this technical solution, since the wiper as a whole moves within the approximate rectangular range formed by the windshield, a Cartesian coordinate system and an angular coordinate system can be established based on the coordinates of the corner points of the four corners of the windshield and the intersection of the wiper and the rectangular area, that is, the coordinates of the first intersection.

[0018] The origin of the Cartesian coordinate system and the pixel length of the unit scale are determined by the four corner points of the windshield. The origin of the angular coordinate system is the root node of the wiper, that is, the first intersection. At this time, the first coordinate (x1, y1) and the second coordinate (x2, y2) of the second intersection and the first endpoint in the Cartesian coordinate system are determined by the vehicle recognition model, and the angle alph between the wiper arm and the bottom of the windshield is calculated. At this time, the position of the wiper can be expressed as: (alph, x1, y1, x2, y2), that is, the position information of the wiper.

[0019] In any of the above technical solutions, the vehicle detection method also includes: obtaining first position information corresponding to the first frame image in the video information, and obtaining second position information corresponding to other frame images in the video information; determining a first difference between any second position information and the first position information; based on the existence of at least two first differences that are unequal to each other and any first difference that is greater than a preset threshold, determining that the position information meets the preset conditions.

[0020] In this technical solution, firstly, the wiper position information corresponding to the first frame image in the video information is obtained, that is, the first position information. Then, the wiper position information corresponding to all the frame images in the video information except the first frame image is obtained, that is, the second position information, and the difference between the first position information and the second position information is calculated, which is recorded as the first difference.

[0021] If the first difference is a variable value, that is, at least two first differences are not equal, it means that the wiper is at different positions in at least three frames of images, and it can be determined that the wiper is moving. At the same time, among the multiple first differences, there is at least one first difference greater than a preset threshold, which means that the movement amplitude of the wiper meets the minimum requirement. At this time, it can be determined that the vehicle's wiper is moving normally, that is, the position information meets the preset conditions, and the current vehicle can pass the verification.

[0022] In any of the above technical solutions, before the step of determining the vehicle image information in any frame image in the video information through the vehicle detection model, the vehicle detection method also includes: obtaining a preset detection model and obtaining preset vehicle image information; generating a training set based on the preset vehicle image information, and training the preset detection model through the training set to obtain a vehicle detection model.

[0023] In this technical solution, during the vehicle detection process, a vehicle detection model is needed to identify the vehicle image and the wiper position, so it is necessary to pre-train a vehicle detection model with accuracy that meets the requirements. Specifically, first set an untrained preset detection model, and specifically select a CNN (Convolutional Neural Networks) model, and generate a training set through the labeled preset vehicle image information to train the preset detection model. Among them, the preset vehicle image information includes images of vehicles from multiple angles and multiple brands and models, and in these images, the vehicle wipers are in different positions, and the position information of the wipers is pre-labeled. Through training and learning, the obtained vehicle detection model can accurately identify the position of the wipers in the vehicle image, and then determine whether the wipers move as required based on the position information of the wipers, thereby completing the detection of the vehicle.

[0024] In any of the above technical solutions, the step of training the detection model through the training set specifically includes: inputting the training set into the preset detection model, and obtaining the loss value of the preset detection model, until it is determined that the loss value is less than the preset loss value, and the vehicle detection model is determined to be obtained.

[0025] In this technical solution, after the training set input value is preset into the detection model, there will be a difference between the output value of the detection model and the labeled value of the training set. This difference is the loss value of the current preset detection model, which represents the deviation or loss between the prediction of the wiper position information by the preset detection model and the actual position information. When the loss value is reduced to a certain extent, specifically when the loss value of the detection model is less than the preset loss value, it means that the detection accuracy of the preset detection model after training has met the requirements, and the trained preset detection model can be used as a vehicle detection model.

[0026] In any of the above technical solutions, the loss value is the sum of the first loss value and the second loss value, and the preset vehicle image information includes preset wiper position information; wherein the difference between the predicted position information output by the preset detection model and the preset wiper position information is the first loss value, and the second loss value is calculated by the loss function.

[0027] In this technical solution, in order to improve the speed and accuracy of training and make the vehicle detection model easier to optimize, the present invention proposes that the loss value includes a first loss value and a second loss value, wherein the first loss value is the difference between the output value of the preset model and the labeled value of the training set, and this part of the loss value is determined by the manually calibrated key point coordinates and the predicted key point coordinates. The second loss value is calculated based on a loss function with a self-supervisory effect. When the sum of the first loss value and the second loss value is less than the preset loss value, it is determined that the trained preset detection model can be used as a vehicle detection model.

[0028] In any of the above technical solutions, the third coordinate corresponding to the second endpoint of the brush body is determined by a preset detection model; the loss function is specifically:

[0029] L self =‖(y2-y1)(x3-x1)-(y3-y1)(x2-x1)‖;

[0030] Among them, L self is the second loss value, (x1, y1) is the first coordinate, (x2, y2) is the second coordinate, and (x3, y3) is the third coordinate.

[0031] In this technical solution, the third coordinate corresponding to the second endpoint of the brush body is obtained. Since the first endpoint of the brush body, the second endpoint of the brush body, and the first intersection of the brush body and the rotating arm are all located on the brush body, these three points are approximately on the same straight line. The present invention relates to the above-mentioned loss function, which calculates the second loss value according to the first coordinate (x1, y1), the second coordinate (x2, y2) and the third coordinate (x3, y3) corresponding to the first endpoint, the first intersection and the second endpoint. In the specific formula, by finding the norm of the first coordinate, the second coordinate and the third coordinate, it is finally determined whether the above three key points are on the same straight line. If the three key points are close to the same straight line, the second loss value Lself approaches zero.

[0032] The second aspect of the present invention provides a vehicle detection system, comprising: a memory configured to store a computer program; a processor configured to execute the computer program to implement the vehicle detection method provided in any of the above technical solutions. Therefore, the vehicle detection system includes all the beneficial effects of the vehicle detection method provided in any of the above technical solutions, which will not be repeated here.

[0033] The third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the vehicle detection method provided in any of the above technical solutions is implemented. Therefore, the computer-readable storage medium includes all the beneficial effects of the vehicle detection method provided in any of the above technical solutions, which will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0035] Figure 1 A flow chart of a vehicle detection method according to an embodiment of the present invention is shown;

[0036] Figure 2 Another flow chart of a vehicle detection method according to an embodiment of the present invention is shown;

[0037] Figure 3 Another flow chart of a vehicle detection method according to an embodiment of the present invention is shown;

[0038] Figure 4 A schematic diagram showing windshield wiper position recognition in a vehicle detection method according to an embodiment of the present invention is shown;

[0039] Figure 5 Another schematic diagram showing wiper position recognition in a vehicle detection method according to an embodiment of the present invention;

[0040] Figure 6 A schematic diagram of the structure of a vehicle detection model in a vehicle detection method according to an embodiment of the present invention is shown;

[0041] Figure 7 A flowchart for detecting whether a vehicle is a real vehicle in one embodiment of the present invention is shown;

[0042] Figure 8 A schematic diagram showing a frame image of an original video in one embodiment of the present invention is shown;

[0043] Fig. 9 A schematic diagram showing vehicle image information after being identified by a vehicle detection model in one embodiment of the present invention;

[0044] Fig.10 A structural block diagram of a vehicle detection system according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0045] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0046] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0047] Refer to the following Figures 1 to 10 The vehicle detection method, vehicle detection system and computer-readable storage medium according to some embodiments of the present invention are described.

[0048] Embodiment 1

[0049] like Figure 1 As shown, in an embodiment of the first aspect of the present invention, a vehicle detection method is provided, comprising:

[0050] Step S102, receiving video information;

[0051] Step S104, determining the vehicle image information corresponding to any frame image in the video information through the vehicle detection model, and determining the position information of the wiper of the vehicle to be detected according to the vehicle image information;

[0052] Step S106, determining whether the vehicle to be detected has passed the detection based on whether the position information meets the preset conditions.

[0053] In this embodiment, when the vehicle is detected, the real video information uploaded by the driver is first received. When the driver is asked to shoot the vehicle video, the wipers are turned on at the same time. If the wipers are detected to be operating normally in the video uploaded by the driver, the video uploaded by the driver is considered to be a real vehicle video shot in real time. If the wipers are detected not to be operating normally in the video uploaded by the driver, it indicates that the driver is likely to cheat, and the vehicle in the uploaded video is not a real "live" vehicle, and the vehicle detection fails at this time.

[0054] Specifically, the received video information is input into the vehicle detection model, and the video information is processed by the vehicle detection model to obtain continuous frame images, and each frame image is further processed into corresponding vehicle image information, and the position information of the vehicle's windshield wiper in the vehicle image information is determined frame by frame. When the windshield wiper meets the preset conditions built into the vehicle detection model, it is determined that the windshield wiper operates normally as required, and at this time it is determined that the vehicle to be detected has passed the detection.

[0055] The embodiment provided by the present invention determines whether the vehicle video uploaded by the driver is a real "live" vehicle video according to the movement state of the wiper when detecting the vehicle video uploaded by the driver. It can effectively distinguish cheating methods such as uploading old videos and reshooting the screen, thereby improving the accuracy of video detection.

[0056] In one embodiment of the present invention, the wiper includes a rotating arm and a brush body, and the vehicle to be detected also includes a front windshield; Figure 2 As shown, the step of determining the position information of the wiper of the vehicle to be detected according to the vehicle image information specifically includes:

[0057] Step S202, determining a first intersection point of a side where the rotating arm intersects with the front windshield through a vehicle detection model, determining a second intersection point of the rotating arm and the brush body, and determining a first end point of the brush body;

[0058] Step S204: determining position information according to the first intersection point, the second intersection point and the first endpoint.

[0059] In this embodiment, the wiper of a general vehicle includes a rotating arm and a brush body. After being identified by the vehicle detection model, the front windshield of the vehicle is identified as an approximately rectangular area, and the wiper moves within the rectangular area. Specifically, the point where the wiper intersects with the bottom edge of the rectangular area corresponding to the front windshield is the root node of the wiper, that is, the first intersection point, and the wiper rotates approximately around the root node. The position where the rotating arm of the wiper intersects with the brush body is the second intersection point, and the brush body rotates relative to the rotating arm around the second intersection point. At the same time, the brush body itself is processed into a shape that approximates a line segment, and its two ends are marked as the first endpoint and the second endpoint respectively.

[0060] It can be obtained that the first intersection and the second intersection are on the same straight line, that is, they are on the same arm of the wiper. The second intersection and the first endpoint are on the same straight line, that is, the brush body of the wiper. Therefore, the positions of the wiper arm and the brush body can be determined through the first intersection, the second intersection and the first endpoint, and then the position information of the wiper can be obtained.

[0061] In one embodiment of the present invention, the step of determining the position information based on the first intersection, the second intersection and the first endpoint specifically includes: determining the first corner point, the second corner point, the third corner point and the fourth corner point corresponding to the front windshield through a vehicle detection model; establishing a Cartesian coordinate system with any one of the first corner point, the second corner point, the third corner point and the fourth corner point as the origin, and establishing an angular coordinate system with the first intersection as the origin; determining the angle data between the side where the front windshield intersects with the swing arm and the swing arm in the angular coordinate system; determining the first coordinate corresponding to the first endpoint and the second coordinate corresponding to the second intersection in the Cartesian coordinate system; determining the position information based on the angle data, the first coordinate and the second coordinate.

[0062] In this embodiment, since the wiper as a whole moves within the approximate rectangular range formed by the windshield, a Cartesian coordinate system and an angular coordinate system can be established based on the coordinates of the corner points of the four corners of the windshield and the intersection of the wiper and the rectangular area, that is, the coordinates of the first intersection.

[0063] The origin of the Cartesian coordinate system and the pixel length of the unit scale are determined by the four corner points of the windshield. The origin of the angular coordinate system is the root node of the wiper, that is, the first intersection. At this time, the first coordinate (x1, y1) and the second coordinate (x2, y2) of the second intersection and the first endpoint in the Cartesian coordinate system are determined by the vehicle recognition model, and the angle alph between the wiper arm and the bottom of the windshield is calculated. At this time, the position of the wiper can be expressed as: (alph, x1, y1, x2, y2), that is, the position information of the wiper.

[0064] In one embodiment of the present invention, Figure 3 As shown, the vehicle detection method also includes:

[0065] Step S302, obtaining first position information corresponding to a first frame image in the video information, and obtaining second position information corresponding to other frame images in the video information;

[0066] Step S304, determining a first difference between any second position information and the first position information;

[0067] Step S306: determining that the location information meets a preset condition based on the existence of at least two first differences that are not equal to each other and any first difference being greater than a preset threshold.

[0068] In this embodiment, firstly, the wiper position information corresponding to the first frame image in the video information is obtained, that is, the first position information. Then, the wiper position information corresponding to all the frame images in the video information except the first frame image is obtained, that is, the second position information, and the difference between the first position information and the second position information is calculated, which is recorded as the first difference.

[0069] If the first difference is a variable value, that is, at least two first differences are not equal, it means that the wiper is at different positions in at least three frames of images, and it can be determined that the wiper is moving. At the same time, among the multiple first differences, there is at least one first difference greater than a preset threshold, which means that the movement amplitude of the wiper meets the minimum requirement. At this time, it can be determined that the vehicle's wiper is moving normally, that is, the position information meets the preset conditions, and the current vehicle can pass the verification.

[0070] In one embodiment of the present invention, before the step of determining the vehicle image information in any frame image in the video information through the vehicle detection model, the vehicle detection method also includes: obtaining a preset detection model and obtaining preset vehicle image information; generating a training set based on the preset vehicle image information, and training the preset detection model through the training set to obtain a vehicle detection model.

[0071] In this embodiment, during the vehicle detection process, a vehicle detection model is required to identify the vehicle image and the wiper position, so it is necessary to pre-train a vehicle detection model with accuracy that meets the requirements. Specifically, an untrained preset detection model is first set, and a CNN (Convolutional Neural Networks) model can be selected, and a training set is generated through the labeled preset vehicle image information to train the preset detection model. Among them, the preset vehicle image information includes images of vehicles from multiple angles and multiple brands and models, and in these images, the vehicle wipers are in different positions, and the position information of the wipers is pre-labeled. Through training and learning, the obtained vehicle detection model can accurately identify the position of the wipers in the vehicle image, and then determine whether the wipers move as required based on the position information of the wipers, thereby completing the detection of the vehicle.

[0072] In one embodiment of the present invention, the step of training the detection model through the training set specifically includes: inputting the training set into the preset detection model, and obtaining the loss value of the preset detection model, until it is determined that the loss value is less than the preset loss value, and the vehicle detection model is determined to be obtained.

[0073] In this embodiment, after the training set input value is preset to the detection model, there is a difference between the detection model output value and the labeled value of the training set. This difference is the loss value of the current preset detection model, which represents the deviation or loss between the prediction of the wiper position information by the preset detection model and the actual position information. When the loss value is reduced to a certain extent, specifically when the loss value of the detection model is less than the preset loss value, it means that the detection accuracy of the preset detection model after training has met the requirements, and the trained preset detection model can be used as a vehicle detection model.

[0074] In one embodiment of the present invention, the loss value is the sum of a first loss value and a second loss value, and the preset vehicle image information includes preset wiper position information; wherein the difference between the predicted position information output by the preset detection model and the preset wiper position information is the first loss value, and the second loss value is calculated by the loss function.

[0075] In this embodiment, in order to improve the speed and accuracy of training and make the vehicle detection model easier to optimize, the present invention proposes that the loss value includes a first loss value and a second loss value, wherein the first loss value is the difference between the output value of the preset model and the labeled value of the training set, and this part of the loss value is determined by the manually calibrated key point coordinates and the predicted key point coordinates. The second loss value is calculated based on a loss function with a self-supervisory effect. When the sum of the first loss value and the second loss value is less than the preset loss value, it is determined that the trained preset detection model can be used as a vehicle detection model.

[0076] In one embodiment of the present invention, the third coordinate corresponding to the second end point of the brush body is determined by a preset detection model; the loss function is specifically:

[0077] L self =‖(y2-y1)(x3-x1)-(y3-y1)(x2-x1)‖;

[0078] Among them, L self is the second loss value, (x1, y1) is the first coordinate, (x2, y2) is the second coordinate, and (x3, y3) is the third coordinate.

[0079] In this embodiment, the third coordinate corresponding to the second endpoint of the brush body is obtained. Since the first endpoint of the brush body, the second endpoint of the brush body, and the first intersection of the brush body and the rotating arm are all located on the brush body, these three points are approximately on the same straight line. The present invention relates to the above-mentioned loss function, and the second loss value is calculated according to the first coordinate (x1, y1), the second coordinate (x2, y2) and the third coordinate (x3, y3) corresponding to the first endpoint, the first intersection and the second endpoint. In the specific formula, by finding the norm of the first coordinate, the second coordinate and the third coordinate, it is finally determined whether the above three key points are on the same straight line. If the three key points are close to the same straight line, the second loss value Lself approaches zero.

[0080] Embodiment 2

[0081] In one embodiment of the present invention, a complete detection process of whether a vehicle is a live vehicle is taken as an example to illustrate the embodiment of the present invention:

[0082] First, the concept of live vehicle detection is explained. A similar concept is face liveness detection. There are currently two solutions for face liveness detection, interactive liveness detection and silent liveness detection. Interactive face liveness detection completes face liveness authentication by detecting whether the face blinks, opens the mouth, shakes the head, etc.; silent face liveness detection relies on a large amount of real face data and non-real face data, and uses the model to learn the features of non-live faces and live faces to complete face liveness authentication.

[0083] The live vehicle detection has a similar idea. The present invention proposes to perform live authentication on the vehicle based on whether the change in wiper position meets the requirements, so as to verify whether the vehicle is live.

[0084] Specifically, a Cartesian coordinate system and an angular coordinate system are established respectively through the four corner points of the front windshield of the vehicle and the root node of the wiper, wherein the origin of the Cartesian coordinate system and the pixel length of the unit scale are determined by the four corner points of the window, and the origin of the angular coordinate system is the root node of the wiper.

[0085] Specific as Figure 4 As shown, the wiper position can be expressed as (alph, x1, y1, x2, y2).

[0086] At the same time, if Figure 5 and Figure 6 As shown, the present invention adopts a three-stage cascade CNN key point detection model to detect the four corner points of the front window, the root node of the wiper, and the three key points of the wiper to determine the position of the wiper.

[0087] The first-level CNN model in the three-level CNN model uses a fully convolutional network, inputs a vehicle image, and outputs the window area, the coordinates of the four window corner points, and the root node coordinates of the wiper.

[0088] The input of the second-level CNN model is the window area output by the first-level CNN model, and the output is the coordinates of the window corner points and the root node of the wiper.

[0089] The input of the third-level CNN model is the window corner points output by the second-level CNN model, the aligned window area determined by the root node of the wiper, and the output is the three key points of the wiper.

[0090] The entire three-level cascade CNN key point detection model detects image information from coarse to fine, from easy to difficult. The first-level CNN model detects coarse information of the window area, and the second-level CNN model detects fine information such as the window corners and the wiper root nodes. The third-level CNN model detects fine information such as the three key points of the wiper. The three-level CNN cascade network decouples the detection of fixed-position key points such as window corners and wiper root nodes from non-fixed key points such as wiper key points, making the model easier to optimize.

[0091] Among them, corner regression is also added to the first-level CNN model to assist model training and give the model more supervision information to obtain better detection performance.

[0092] The present invention also designs a loss function with self-supervision to make the wiper position detection model easier to optimize. Specifically, no matter how the wiper position changes, the positions of the three key points on the wiper are always approximately on the same straight line. The present invention adds this prior information to the loss function of the third-level CNN to provide a stronger supervision signal. Its definition is as follows:

[0093] L self =‖(y2-y1)(x3-x1)-(y3-y1)(x2-x1)‖;

[0094] Among them, L self is the second loss value, (x1, y1) is the first coordinate, (x2, y2) is the second coordinate, and (x3, y3) is the third coordinate.

[0095] The present invention verifies whether the driver can control the vehicle by verifying whether the driver has turned on the wipers, and then determines whether the vehicle is alive. The platform issues a command to turn on the wipers, and the driver operates according to the command and shoots a video to upload. The present invention detects the wiper position of each frame in the video, and then calculates the distance between the wipers in each frame to determine whether the wipers are turned on. The specific process is as follows Figure 7 As shown:

[0096] Step S702, the platform issues a command to turn on the wipers;

[0097] Step S704: the vehicle verification personnel turns on the wipers according to the instruction and takes a video and uploads it;

[0098] Step S706, using the vehicle detection model to detect the vehicle area and generate a new video stream;

[0099] Step S708, predicting the wiper position of each frame of the image;

[0100] Step S710, calculating the distance of the wiper position of each frame of the image;

[0101] Step S712, determine whether the wiper is turned on; if yes, proceed to step S714, otherwise proceed to step S716;

[0102] Step S714, determining that it is a real vehicle;

[0103] Step S716, determining that it is a non-real vehicle.

[0104] In step S706, the frame image of the original video is as follows: Figure 8 As shown in the figure, the vehicle image information obtained after the vehicle detection model is identified is as follows Fig. 9 The new video stream is input into the key point detection model to detect the four corner points of the front window, the root node of the wiper, and the three key points of the wiper in each frame, and the Cartesian coordinate system and the angular coordinate system are established to calculate the wiper position.

[0105] In step S712, if the calculated distance is a series of changing values ​​and the maximum distance value is greater than the set threshold, the vehicle is considered to be a real vehicle (living object); otherwise, the vehicle is considered to be a non-real vehicle (non-living object).

[0106] Embodiment 3

[0107] like Fig.10 As shown, in one embodiment of the present invention, a vehicle detection system 1000 is provided, comprising: a memory 1002, configured to store a computer program; a processor 1004, configured to execute the computer program to implement the vehicle detection method provided in any of the above embodiments. Therefore, the vehicle detection system 1000 includes all the beneficial effects of the vehicle detection method provided in any of the above embodiments, which will not be repeated here.

[0108] Embodiment 4

[0109] In one embodiment of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the vehicle detection method provided in any of the above embodiments is implemented. Therefore, the computer-readable storage medium includes all the beneficial effects of the vehicle detection method provided in any of the above embodiments, which will not be repeated here.

[0110] In the description of the present invention, the term "plurality" refers to two or more than two. Unless otherwise clearly defined, the orientation or positional relationship indicated by the terms "upper" and "lower" is based on the orientation or positional relationship described in the accompanying drawings. It is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention. The terms "connection", "installation", "fixation", etc. should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be directly connected or indirectly connected through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0111] In the description of the present invention, the description of the terms "one embodiment", "some embodiments", "specific embodiments", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In the present invention, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0112] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A vehicle detection method, It is characterized in that include: Receive video information; Determine the vehicle image information corresponding to any frame image in the video information through the vehicle detection model, and determine the position information of the wiper of the vehicle to be detected according to the vehicle image information; Based on the position information satisfying a preset condition, determining that the vehicle to be detected has passed the detection; Wherein, the wiper includes a rotating arm and a brush body, and the vehicle to be detected also includes a front windshield; the step of determining the position information of the wiper of the vehicle to be detected based on the vehicle image information specifically includes: determining the first intersection point of the side where the rotating arm intersects with the front windshield through the vehicle detection model, determining the second intersection point of the rotating arm and the brush body, and determining the first end point of the brush body; determining the position information based on the first intersection point, the second intersection point and the first end point.

2. The vehicle detection method according to claim 1, It is characterized in that The step of determining the position information according to the first intersection, the second intersection and the first endpoint specifically includes: Determine a first corner point, a second corner point, a third corner point and a fourth corner point corresponding to the front windshield by using the vehicle detection model; Establishing a Cartesian coordinate system with any one of the first corner point, the second corner point, the third corner point, and the fourth corner point as an origin, and establishing an angular coordinate system with the first intersection point as an origin; Determine the angle data between the side where the front windshield intersects with the rotating arm and the rotating arm in the angular coordinate system; Determine, in the Cartesian coordinate system, a first coordinate corresponding to the first endpoint and a second coordinate corresponding to the second intersection point; The position information is determined according to the angle data, the first coordinate and the second coordinate.

3. The vehicle detection method according to claim 2, It is characterized in that Also includes: Acquire first position information corresponding to a first frame image in the video information, and acquire second position information corresponding to other frame images in the video information; Determine a first difference between any one of the second position information and the first position information; Based on the existence of at least two first difference values ​​that are unequal to each other and any one of the first difference values ​​being greater than a preset threshold, it is determined that the location information satisfies the preset condition.

4. The vehicle detection method according to claim 2, It is characterized in that Before the step of determining the vehicle image information in any frame image in the video information by using the vehicle detection model, the vehicle detection method further includes: Obtain a preset detection model and obtain preset vehicle image information; A training set is generated according to the preset vehicle image information, and the preset detection model is trained by the training set to obtain the vehicle detection model.

5. The vehicle detection method according to claim 4, It is characterized in that The step of training the detection model through the training set specifically includes: The training set is input into the preset detection model, and the loss value of the preset detection model is obtained, until it is determined that the loss value is less than the preset loss value, and the vehicle detection model is determined to be obtained.

6. The vehicle detection method according to claim 5, It is characterized in that The loss value is the sum of a first loss value and a second loss value, and the preset vehicle image information includes preset wiper position information; Among them, the difference between the predicted position information output by the preset detection model and the preset wiper position information is the first loss value, and the second loss value is calculated by the loss function.

7. The vehicle detection method according to claim 6, It is characterized in that Also includes: Determine a third coordinate corresponding to the second end point of the brush body by using the preset detection model; The loss function is specifically: <h2 style=";text-align:left;direction:ltr">L<h2 style=";text-align:left;direction:ltr"> self <h2 style=";text-align:left;direction:ltr"> =‖(y2-y1)(x3-x1)-(y3-y1)(x2-x1)‖; Among them, L self is the second loss value, (x1, y1) is the first coordinate, (x2, y2) is the second coordinate, and (x3, y3) is the third coordinate.

8. The vehicle detection method according to claim 1, It is characterized in that The first intersection point and the second intersection point are on the same straight line; the second intersection point and the first endpoint are on the same straight line.

9. A vehicle detection system, It is characterized in that include: a memory configured to store a computer program; A processor is configured to execute the computer program to implement the vehicle detection method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the vehicle detection method according to any one of claims 1 to 8 is implemented.

Citation Information

Patent Citations

  • Vehicle active safety system alarm shielding device

    CN103770780A

  • Detection method for front automobile window top right corner point

    CN103854029A

  • Vehicle Accessory Operation Based on Motion Tracking

    CN105700394A

  • Automatic security-inspection vehicle tracking and locating method and system of open scene

    CN107808392A

  • State marking method and system of target vehicle and computer readable storage medium

    CN109815856A