基于改进概率霍夫变换的车道线快速检测方法

车道线是行车安全的重要参考。为提高无人驾驶行车过程中车道线检测的准确性和实时性,提出一种基于改进概率霍夫变换的车道线快速检测方法。首先对获取的图像进行感兴趣区域提取,根据车道线颜色的特殊性,合理选取三色通道的比值对图片进行灰度化,为增强阈值处理的鲁棒性,采用大津二值化法对灰度图像进行二值化,由于Canny算子具有良好的定位边缘的能力,本次边缘提取算子选取为Canny。接着分别从车道线长度、角度、车体和车道宽度4个方面提出4点约束条件对该算法加以改进,剔除干扰线和伪车道线,最后通过线性回归法拟合出正确车道线。实验结果表明,该算法在快速检测车道线的同时保证了检测的准确率,并将实验结果与其他算法进行比较,证明了该算法的实时性和准确性优于其他算法。

A Fast Lane Line Detection Method Based on Improved Probability Hough Transform

Lane line is an important reference for traffic safety. To improve the accuracy and timeliness of lane line detection in driverless driving,a fast lane line detection method based on improved probabilistic Hough transform is presented. Firstly,the region of interest is extracted from the acquired image,according to the particularity of lane line color,the ratio of three color channels is selected reasonably to conduct gray-scale of the picture. In order to enhance the robustness of threshold processing,Otsu’s threshold algorithm is used to binarize the gray scale image. The Canny operator,which has a great ability to locate the edge,is chosen in this study. Then,the algorithm is improved by putting forward four-point constraint conditions from four aspects of lane line length,angle,vehicle body and lane width to eliminate interference lines and pseudo-lane lines. Finally,the correct lane line is fitted by linear regression method. The experiment shows that the proposed algorithm can detect lane quickly and ensure the accuracy of lane detection. Comparing the experimental results with other algorithms,it is proved that the real-time and accuracy of the proposed algorithm is better than other algorithms.