A highway anti-dazzle plate shaking detection method based on vehicle-mounted monocular vision

Abstract

The present application relates to a kind of highway anti-glare board shaking detection methods based on vehicle-mounted monocular vision, comprising: image acquisition component and position acquisition component are installed on motor vehicle, so that motor vehicle is driven on road, while collecting image information and position information in driving process, to obtain image data with time and position information;The boundary of anti-glare board object is extracted, so as to calculate the relative inclination of anti-glare board in image relative to image coordinate system;By calculating the relative inclination of all anti-glare boards in each image data, the absolute inclination of each anti-glare board is calculated;The absolute inclination of each anti-glare board is compared with the preset threshold value, if it exceeds threshold value, it is determined that the anti-glare board has occurred shaking.Compared with prior art, the present application can realize the automatic detection of road anti-glare board shaking phenomenon, compared with the current manual detection process, efficiency and precision are significantly improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of road disease detection, in particular to a highway anti-glare plate shaking detection system based on vehicle-mounted monocular vision. BACKGROUND

[0002] On urban viaducts and expressways, the coverage of anti-glare plates is becoming larger and larger. However, due to the repeated action of natural wind or vehicle-induced wind load, the fixing screws of the anti-glare plates are prone to loosen, and the anti-glare plates will shake. If not repaired or replaced in time, the anti-glare plates may even fall onto the road, thereby posing a hidden danger to traffic operation safety. At present, manual detection is generally used, but the workload is large, the efficiency is low, the detection frequency is low, and there is no standardized recording means.

[0003] The existing public schemes such as CN 1191185, CN 108774978A, CN 113957816A, CN 111999182A and CN 216012924U are mainly aimed at the design, production, cleaning, fatigue test and strength test of anti-glare plates, and cannot reasonably improve the above-mentioned problems in the present technical scheme. SUMMARY

[0004] The purpose of the present application is to overcome the defects of the prior art and provide a highway anti-glare plate shaking detection system based on vehicle-mounted monocular vision. By quickly shooting the anti-glare plates on the road, automatic detection of the shaking phenomenon of the road anti-glare plates can be realized, and the efficiency and accuracy of the current manual detection process are significantly improved.

[0005] The purpose of the present application can be achieved by the following technical scheme:

[0006] The purpose of the present technical scheme is to provide a highway anti-glare plate shaking detection method based on vehicle-mounted monocular vision, referring to Figure 1 and Figure 2 , comprising the following steps:

[0007] S1: installing the image acquisition component and the position acquisition component on the motor vehicle, so that the image information and the position information during driving can be acquired simultaneously when the motor vehicle drives on the road, thereby obtaining image data with time and position information;

[0008] S2: based on the computer vision algorithm, processing each image data, first extracting the boundary of the anti-glare plate object, and then calculating the relative inclination angle of the anti-glare plate in the image relative to the image coordinate system;

[0009] S3: calculating the absolute inclination angle of each anti-glare plate by calculating the mode of the relative inclination angles of all anti-glare plates in each image data;

[0010] S4: comparing the absolute inclination angle of each anti-glare plate with a preset threshold value, and determining that the anti-glare plate has occurred shaking if the threshold value is exceeded.

[0011] Further, the image acquisition component comprises a monocular vehicle-mounted camera and a first clock.

[0012] Further, the position acquisition component comprises a GPS locator and a second clock.

[0013] Further, in S1, the first processing unit is used to associate the position information obtained by the position acquisition component and the image information obtained by the image acquisition component, so as to obtain image data with time and position information.

[0014] Further, in S1, the image acquisition component acquires images with time stamps according to a preset frame rate.

[0015] The position acquisition component records the longitude and latitude position information of the motor vehicle at each time point.

[0016] Further, in S1, when the time point corresponding to the image with the time stamp is different from the time corresponding to the longitude and latitude position information, the longitude and latitude position information before and after the image acquisition time is interpolated as the geographical position information of the image.

[0017] Further, in S2, the second processing unit is used to extract the boundaries of all anti-glare plate objects in each frame of image, obtain the coordinates of the upper, lower, left and right four boundary lines of each anti-glare plate, and calculate the relative inclination angle of the left and right two longitudinal boundaries of each anti-glare plate in the image coordinate system.

[0018] Further, in S3, the second processing unit is used to calculate the mode of the boundary relative inclination angle of all anti-glare plates in an image, which is used to represent the vertical direction, compare the boundary relative inclination angle of all anti-glare plates in each image with the mode of the boundary relative inclination angle, and calculate the absolute inclination angle of each anti-glare plate in the image coordinate system relative to the vertical direction.

[0019] Further, in S4, the second processing unit is used to compare the absolute inclination angle of each anti-glare plate with a preset threshold value, and determine that the anti-glare plate has occurred shaking if the absolute inclination angle exceeds the preset threshold value, and store the image in which the anti-glare plate is located and the position of the anti-glare plate in the image.

[0020] Further, in S4, the second processing unit is used to comprehensively obtain the anti-glare plate shaking condition detected in all image data with time and position information, so as to obtain a single anti-glare plate shaking detection result data set.

[0021] Further, the first processing unit and the second processing unit can be selected from the current mainstream MCU or FPGA.

[0022] Compared with the prior art, the present application has the following technical advantages:

[0023] 1) The image acquisition component and the position acquisition component in the technical solution can be conveniently installed on a motor vehicle, and through rapid shooting of the anti-glare board on the road, automatic detection of the shaking phenomenon of the road anti-glare board can be realized, and the efficiency and accuracy are significantly improved compared with the current manual detection process.

[0024] 2) The technical solution can comprehensively obtain the shaking condition of the anti-glare board detected in all image data with time and position information to obtain a single anti-glare board shaking detection result data set. The second processing unit stores the anti-glare board shaking detection result data set in the built-in ROM after each anti-glare board shaking detection, and can export it to an external user terminal for reference and analysis, thereby providing technical support for realizing regular anti-glare board maintenance of the road section. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 FIG. 1 is a flowchart of the highway anti-glare board shaking detection method in the technical solution;

[0026] Figure 2 FIG. 2 is a top view schematic diagram of the highway anti-glare board shaking detection scene in the technical solution. DETAILED DESCRIPTION

[0027] The present application will be described in detail below in combination with the drawings and specific embodiments. The structure / module name, control mode, algorithm and other features not explicitly described in the technical solution are considered as common technical features disclosed in the prior art.

[0028] The highway anti-glare board shaking detection method based on vehicle-mounted monocular vision in the technical solution includes the following steps:

[0029] S1: Install the image acquisition component and the position acquisition component on the motor vehicle, so that the image information and the position information in the driving process are collected at the same time when the motor vehicle drives on the road, so as to obtain image data with time and position information. The specific installation position and posture are set according to the external form of the specific motor vehicle.

[0030] The image acquisition component includes a monocular vehicle-mounted camera and a first clock, so that the image acquisition component collects images with time stamps at a preset frame rate; the position acquisition component includes a GPS locator and a second clock, so that the position acquisition component records the latitude and longitude position information of the motor vehicle at each time point. The first clock and the second clock use the current mainstream electronic clock element, and are connected in communication with an external computer terminal, or are connected in communication with the first processing unit and the second processing unit to realize time synchronization, so as to realize the synchronization and adjustment of time.

[0031] The first processing unit is used to associate the position information obtained by the position acquisition component and the image information obtained by the image acquisition component to obtain image data with time and position information.

[0032] When the time point corresponding to the image with the timestamp is different from the time corresponding to the latitude and longitude position information, the latitude and longitude position information before and after the image acquisition time is interpolated as the geographical position information of the image.

[0033] S2: Based on the computer vision algorithm, the boundary of the anti-glare plate object is extracted first, so as to calculate the relative inclination angle of the anti-glare plate in the image relative to the image coordinate system (the program for establishing the image coordinate system is pre-stored in the built-in ROM of the processing unit, and the remaining related programs are also pre-stored for loading in the RAM for execution by the CPU, wherein the programs involved are all taken from the mature open source programs, and will not be described here). Through the second processing unit, the boundary of each anti-glare plate object in each frame of image is extracted to obtain the coordinates of the upper, lower, left and right four boundary lines of each anti-glare plate (which can be the coordinates of multiple points on the boundary line or a direct fitting of the straight line relationship of the boundary line), and the relative inclination angle of the left and right two longitudinal boundaries of each anti-glare plate in the image coordinate system is calculated.

[0034] S3: The absolute inclination angle of each anti-glare plate is calculated by calculating the mode of the relative inclination angle of all anti-glare plates in each image data. Through the second processing unit, the mode of the boundary relative inclination angle of all anti-glare plates in an image is calculated to represent the vertical direction. The absolute inclination angle of each anti-glare plate relative to the vertical direction in the image coordinate system is calculated by comparing the boundary relative inclination angle of all anti-glare plates in each image with the mode of the boundary relative inclination angle.

[0035] S4: The absolute inclination angle of each anti-glare plate is compared with a preset threshold value. If the absolute inclination angle exceeds the preset threshold value, it is determined that the anti-glare plate has shaken. Through the second processing unit, the absolute inclination angle of each anti-glare plate is compared with a preset threshold value. If the absolute inclination angle exceeds the preset threshold value, it is determined that the anti-glare plate has shaken, and the image in which the anti-glare plate is located and the position of the anti-glare plate in the image are stored. Through the second processing unit, the shaking conditions of the anti-glare plates detected in all image data with time and position information are integrated to obtain a single anti-glare plate shaking detection result data set. The anti-glare plate shaking detection result data set is stored in the built-in ROM of the second processing unit after each anti-glare plate shaking detection is executed, and can be exported to an external user terminal for review and analysis.

[0036] In specific implementation, the first processing unit and the second processing unit can be selected from the currently mainstream models of MCU or FPGA to realize the graphic processing task and the information processing task.

[0037] The foregoing description of the embodiments has been presented for the purpose of illustration and description. It is not intended to be exhaustive or to limit the application to the precise form disclosed. Modifications and variations are possible in light of the above teachings or can be acquired from practice of the application. As well, the description is presented in the context of the preferred embodiments as a number of alternatives. It is not intended to limit the application to the precise form described.

Claims

1. A highway anti-dazzle board shaking detection method based on vehicle-mounted monocular vision, characterized in that, The method comprises the following steps: S1: mounting an image acquisition component and a position acquisition component on a motor vehicle, so that when the motor vehicle is driving on a road, image information and position information during driving are simultaneously acquired to obtain image data with time and position information; S2: based on a computer vision algorithm, processing each image data, first extracting the boundary of the anti-dazzle plate object, and then calculating the relative inclination angle of the anti-dazzle plate in the image relative to the image coordinate system; S3: calculating the absolute inclination angle of each anti-dazzle plate by calculating the mode of the relative inclination angles of all anti-dazzle plates in each image data; S4: comparing the absolute inclination angle of each anti-dazzle plate with a preset threshold value, and if the threshold value is exceeded, it is determined that the anti-dazzle plate has shaken; In S2, the second processing unit is used to extract the boundary of each anti-dazzle plate object in each frame of image, obtain the coordinates of the upper, lower, left and right four boundary lines of each anti-dazzle plate, and calculate the relative inclination angle of the left and right two longitudinal boundaries of each anti-dazzle plate in the image coordinate system; In S3, the second processing unit is used to calculate the mode of the boundary relative inclination angles of all anti-dazzle plates in an image, which is used to represent the vertical direction, and compare the boundary relative inclination angles of all anti-dazzle plates in each image with the mode of the boundary relative inclination angles, and calculate the absolute inclination angle of each anti-dazzle plate in the image coordinate system relative to the vertical direction.

2. The highway anti-dazzle board shaking detection method based on vehicle-mounted monocular vision according to claim 1, characterized in that, The image acquisition component comprises a monocular vehicle-mounted camera and a first clock. 3.The highway anti-glare board shaking detection method based on vehicle-mounted monocular vision of claim 2, characterized in that, The position acquisition component comprises a GPS locator and a second clock.

4. The highway anti-glare board shaking detection method based on vehicle-mounted monocular vision according to claim 3, characterized in that, In S1, the first processing unit is used to associate the position information obtained by the position acquisition component with the image information obtained by the image acquisition component to obtain image data with time and position information.

5. The highway anti-glare board shaking detection method based on vehicle-mounted monocular vision according to claim 4, characterized in that, In S1, the image acquisition component acquires images with time stamps at a preset frame rate. The position acquisition component records the latitude and longitude position information of the motor vehicle at each time point.

6. The highway anti-glare board shaking detection method based on vehicle-mounted monocular vision according to claim 5, characterized in that, In S1, when the time point corresponding to the image with the time stamp is different from the time corresponding to the latitude and longitude position information, the latitude and longitude position information before and after the image acquisition time is interpolated as the geographical position information of the image.

7. The vehicle-mounted monocular vision-based highway anti-glare board shaking detection method according to claim 1, characterized in that, In S4, the second processing unit is used to compare the absolute inclination angle of each anti-dazzle plate with a preset threshold value, and if the absolute inclination angle exceeds the preset threshold value, it is determined that the anti-dazzle plate has shaken, and the image in which the anti-dazzle plate is located and the position of the anti-dazzle plate in the image are stored. 8.The highway anti-glare board shaking detection method based on vehicle-mounted monocular vision of claim 7, wherein, In S4, the second processing unit is used to comprehensively analyze the shaking conditions of the anti-dazzle plates detected in all image data with time and position information to obtain a single anti-dazzle plate shaking detection result data set.

Images