Research On Road Sag Detection Technology Based On Line Structured Light

The measurement technology based on structured light has the advantages of noncontact,high accuracy,good real-time,active control and low cost,and is widely used in industrial detection.The use of structured light detection technology for road obstacle or depression detection can provide real-time road information for vehicles or mobile platforms and be used to support autonomous driving.Based on the line structure optical vision measurement system and the distance and morph measurement of the front sag as the goal,this thesis carries out the extraction of the system calibration,the calculation of the center distance and other related studies,mainly including the following parts:(1)Selection of the measuring system equipment.According to the actual requirements of the line structure optical vision measurement system,build the industrial camera,computer,one-line linear laser and motion trolley hardware with appropriate performance,and design the measurement range of the actual measurement system and explain it by adjusting the parameters and functions of the equipment.The bracket arm is fixed in the camera position,and the laser transmitter is used to project from multiple angles so that the camera collects the image with multiline structure light stripes,and processes the coordinate conversion relationship between two-dimensional and three-dimensional space to obtain the data.(2)System parameter calibration.Analyzing the advantages and disadvantages of camera calibration based on two-dimensional checkerboard targets,using Zhang Zhengyou calibration method,calibrating the camera by using the angular features of the twodimensional checkerboard calibration board,and solving the camera parameters.On the basis of the existing light plane calibration,the parameters of the light plane equation are solved according to the intersection feature point of the light bar and the calibration plate at different heights of the calibration plate.At the same time,the characteristic formula for the surface information of the depressed object.(3)Multiline light bar center extraction.This thesis combines the Zhang-sun refinement algorithm and the direction template method.The algorithm in the light image of RGB color space extraction area of interest,and using Gaussian filtering to denoising and morphological processing,and then use the center pixel of adjacent eight neighborhood direction implementation algorithm to reduce the calculation of the longitudinal direction of the final in the normal direction using the direction template adaptation can not only accurately extract the central pixels,can also to fill the vacancy light small gap,maintain stability.(4)Distance at road depression.The actual measurement device platform is built to experiment on the depression environment.According to the deformation information of the light strip,the depression Angle is obtained by the relationship between the average longitudinal distance difference in the depression and the flat road and the Angle,and finally the derived information solution formula is used to solve the pose distance.The error analysis of the experimental data is conducted,and the average error proves the feasibility of the system measurement,which has certain practicability in the direction of vehicle deceleration and avoidance disposal.