基于改进Radon变换的直线钢轨识别算法

针对传统Radon变换在直线轨道检测方面速度慢且不能从众多边缘直线中识别出钢轨,提出基于改进Radon变换的直线轨道识别算法。首先,通过阈值筛选的边缘点;其次,利用先验知识,建立极角约束域;然后通过Radon变换提取所有直线,再利用直线角度差和直线平均距离筛选出钢轨边缘直线;最后将本文算法与传统Hough变换算法和传统Radon变换算法的比较,实验结果显示本文算法能精准识别出钢轨,且速度更快。

Linear Rail Recognition Algorithm Based on Improved Radon Transform

Because of the slow detection speed and the failure of the tradition Radon transform to recognize the rail from multi edge lines,a new and improved method is proposed based on Radon transform. Firstly,edge points are selected based on threshold; secondly,prior knowledge is used to establish a polar angle constraint area; thirdly the Radon transform is employed to extract edge lines; then the angle difference and average distance are utilized to identify the edge line of the rail; finally,the algorithm is compared with the traditional algorithm Hough transform and traditional Radon transform algorithm. The experiment results show that improved Radon algorithm is faster and accurate in recognizing the rail.