摘要
小车在平稳路面行驶,同时前方无车辆障碍,基于此,本研究设计了长距离车道检测系统。在识别车道线时,利用输入的视频进行图像预处理,提取感兴趣区域,获取车道线的最大区域,对其进行透视变换和标定,使用Sobel算子对其二值化,使用滑块法拟合车道线并标定,输出检测后的标定视频文件。该系统能有效检测更远距离的车道线,可以应用于快速行驶路段与有清晰车道线的高速公路。
The car is running on a smooth road with no vehicle obstacles in front. Based on this, this study designed a long-distance lane detection system. When the lane line is identified, the input video is used for image preprocessing, the region of interest is extracted, the maximum area of lane line is obtained, perspective transformation and calibration are carried out, binarization is conducted with Sobel operator, lane line is fitted and calibrated with slider method, and the calibration video file after detection is output. This system can effectively detect lane lines at longer distances and can be applied to fast-moving road sections and highways with clear lane lines.
作者
马泉钧
何自超
林邦演
曾文轩
MA Quanjun;HE Zichao;LIN Bangyan;ZENG Wenxuan(Guangzhou College of South China University of Technology,Guangzhou Guangdong 510800)
出处
《河南科技》
2019年第29期111-113,共3页
Henan Science and Technology
关键词
车道线检测
边缘检测
智能交通
长距离检测
透视变换
lane detection
edge detection
intelligent transportation
long-range detection
perspective transformation
