基于相位特征增强的低对比度磨损车道标线提取方法  

Extraction Method of Low-contrast Worn Lane Markings Based on Phase Feature Enhancement

作　　者：柴辉照 CHAI Huizhao(Shanxi Intelligent Transportation Research Institute Co.,Ltd.,Taiyuan,Shanxi 030032,China)

机构地区：[1]山西省智慧交通研究院有限公司,山西太原030032 [2]智慧交通山西省实验室,山西太原030032

出　　处：《山西交通科技》2023年第5期111-114,119,共5页Shanxi Science & Technology of Transportation

基　　金：山西交通科学研究院集团有限公司创新发展计划揭榜项目(21-JKCF-55);山西省科技厅青年基金项目(202103021223464);山西省科技厅面上项目(202303021211340)。

摘　　要：为解决磨损车道标线在光照不均、对比度低情况下检测与识别不完整问题,通过一种基于相位特征增强的低对比度磨损车道标线识别方法,研究了低对比度下磨损车道标线在部分直线不清晰下的相位特征提取弱能量信息,以利于Hough变换提取标线的直线信息。结果表明:提出方法对光照不均匀、低对比度且磨损情况下的车道标线边缘提取效果较好。可见对磨损或有缺失的车道线标线解析其相位特征并结合Hough变换的鲁棒性能解决车道线检测与识别的不完整问题。To solve the problems of the worn lane markings such as uncomplete detection and identification under uneven illumination and low contrast,according to an identification method for the low-contrast worn lane markings based on the phase feature enhancement,this paper studied extracting weak energy information from phase feature of worn lane markings under low contrast conditions when several lines were not clear,so as to extract line information by Hough transform.The results showed that the proposed method was effective in extracting the lane marking edge under uneven illumination,low contrast,and abrasion,indicating that analyzing the phase feature of worn or missing lane markings and combining the robustness of Hough transform can solve the problems of uncomplete detection and identification of lane markings.

关 键 词：车道标线 低对比度 相位特征 HOUGH变换 

分 类 号：U491.523[交通运输工程—交通运输规划与管理]