Robustness Improvement Method of Lane Detection Based on Optimized Recognizable ROI Zone
Lane detection is the key technology to realize autonomous driving,however the traditional lane detection method is af-fected by the non-lane line with complex background information,resulting in the poor lane detection robustness and recognition rate.In order to improve the robustness of lane detection,an improved method of lane detection based on optimized recognizable ROI zone is proposed in this paper.It is achieved through dynamically identifying the ROI zone,eliminating the image RGB information outside the lane,as well as limiting Hough peak detection space.The verified results show that the method can effectively filter the influence of the non-lane straight line in the background image,thus improving the robustness of lane detection.
lane detection robustness optimized recognizable zone image RGB information hough transform
Liu Xiaonan Chen Wenjin Li Weihua Liu Jie
Dongfeng Nissan Passenger Vehicle Company;South China University of Technology Dongfeng Nissan Passenger Vehicle Company South China University of Technology
国内会议
上海
英文
210-214
2017-10-24（万方平台首次上网日期，不代表论文的发表时间）