基于改进概率霍夫变换算法的车道检测方法研究

Research of Lane Detection Based on Progressive Probabilistic Hough Transform

为了提高车道检测的实时性,提出了一种基于改进的概率霍夫变换算法的车道检测方法。改进的概率霍夫变换算法的特点是用设定累加器阈值的方法减少了检测直线的运算量,通过随机取点映射到参数空间使最长直线被最早检测到的概率最大,比现有车道检测方法中普遍采用的标准霍夫变换算法更具实时性。实际路面图像的对比处理结果证明了该检测方法的实时性、精确性和鲁棒性。

To improve the real - time ability of lane detection, this paper presents a lane detection method based on progressive probability Hough transform. The progressive probability Hough transform sets an accumulator threshold to reduce the calculation quantity of line exaction, selects edge - point randomly so that the most obvious line would be detected first usually. It has better real- time ability than standard Hough transform, which is used commonly in existing lane detection methods. The experiment results of lane detection from images of real road scenes show the accuracy, robustness and real - time ability of the proposed method.