摘要
夜间车道检测是汽车防撞系统适合在夜间工作的前提.分析了不同照明条件下夜间车道图像的特点,介绍了夜间车道的反射成像模型.根据模型特点,提出了一种基于光密度差的对数Prewitt边缘检测和Hough变换的车道检测算法.算法实验结果表明:该算法比白天典型的车道检测算法具有更好的夜间车道检测能力.
Nighttime lane detection was the premise for vehicle anti-collision alarm system working at night. Based on an analysis of the characteristics of the nighttime lane in different illuminationenvironment, the model of reflection image for nighttime lane was introduced. According to the model, an algorithm based on logarithmic Prewitt edged detection of the difference of luminous density and Hough transform was proposed. The results of experiment showed that the proposed algorithm demonstrates better performance for nighttime environment than the traditional daytime algorithm.
出处
《安徽大学学报（自然科学版）》
CAS
北大核心
2008年第6期44-47,共4页
Journal of Anhui University(Natural Science Edition)
关键词
复杂环境
夜间车道检测
光密度
HOUGH变换
complex environment
nighttime lane detection
luminous density
Hough transform
