车道检测方法综述

A Review of the Lane Detection Methods

车道检测是驾驶辅助系统中提高驾驶安全性的重要因素,因此已成为当下驾驶辅助系统领域的研究热点。文章介绍了常见的车道线检测方法,包括基于边缘检测算法和高斯平滑算法的图像预处理,基于线检测,线过滤和聚类的方式和基于视觉的两种不同车道线检测方法。结果表明,在各种照明条件下的平均检测率为93%,执行时间为33 ms,对于实时应用来说足够快。

Lane detection is an important factor to improve driving safety in driving assistance system,so it has become a hot research topic in the field of driving assistance system.This paper introduces common lane detection methods based on vision,including image preprocessing based on edge detection and Gaussian smoothing algorithms,line detection,line filtering and clustering methods-based,and color-based lane line detection methods.Results show that the average detection rate is 93%and an execution time is 33 ms,are fast enough for real-time applications.