基于数字图像处理的直线型铁轨的自动识别

在铁路交通安全检测系统中,铁轨的识别是障碍检测物的关键。提出适合轨道数字图像的直线型铁轨自动识别算法。首先基于轨道图像的亮度特征和直线特征,综合应用Hough变换和边缘提取算法,检测出大量离散的共线像素点。然后建立直线方程,使检测到的直线连续化,根据线段斜率特性和线段长短滤除多余的线段。最后根据提取到的铁轨边缘轮廓得到图像中的铁轨位置信息,为铁路图像信息提取分析打下基础。用MATLAB编程实现了该算法,通过大量仿真实验验证了该算法的有效性。

Based on digital image processing of automatic identify of linear tracks

The recognition of tracks is the key of obstacle detection in the railway transport safety inspection system. In this paper proposed an automatic identification algorithm of linear tracks to suitable for the rails digital image. Firstly of all, the article based on the brightness and linear features of the image, detecting a large number of discrete collinear pixels that make full use of Hough transform and edge extraction algorithm. Secondly, building linear equation to make the detected linear continuous, and filter out redundant line according to the line slope and the length features. Finally, getting the location information of tracks according to extract the edge contour track, lays the foundation for further railway image analysis. At the same time, the paper using Matlab programming realized the algorithm, and through a large number of simulation experiments verify the effectiveness of the proposed algorithm.