基于线结构光路面车辙检测系统研究

Research on rutting detection system based on linear structure light pavement

针对二维图像测量路面车辙时复杂背景干扰的弊端,提出了基于局部逼近式的线结构光路面车辙测量的方法。检测车系统在自然驾驶过程中,向路面车辙投射红外线激光,并通过高速相机采集拍摄完成的图像数据,采用相应的图像处理的算法实现对路面车辙的实时检测。首先介绍了路面车辙测量的三角形原理和线结构光条中心的不同提取方法,然后提出基于局部逼近式的车辙轮廓线深度信息的提取方法,并对比给出了测量不同类型车辙的深度信息。结果表明基于线结构光测量结果较传统人工测量车辙结果精度提高了3%以上,且对于不同类型车辙其测量结果提高各异,突显了线结构光法测量的准确性和差异性。

Due to the disadvantages of complex background interference in two-dimensional image measurement of pavement rutting,a research method based on local approximation for line structure eccentricity measurement system is proposed.The detecting vehicle system project infrared laser light onto the pavement rut during natural driving,and capture the completed image data through a high-speed camera.At last,corresponding image processing algorithms are adopted to real-time detection of pavement This paper first gives the triangular principle of road rutting measurement and different extraction methods of line structure light bar center.Then,a line depth information extraction method based on local approximation is proposed.And give specific information to measure the depth of different types of ruts.The results show that the accuracy of the line-based light measurement is more than 3%higher than that of the traditional manual measurement rutting.And for various types of ruts,the measurement results are improved differently.The above highlights the accuracy and variability of line-structured light measurement.