http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
스테레오 비전을 이용한 실시간 차선 감지 알고리듬 개발
박재학(Jaehak Park),이광운(Kwangwoon Rhee),박종웅(Jongwoung Park),허건수(Kunsoo Huh),이경수(Kyongsu Yi) 한국자동차공학회 2003 한국자동차공학회 춘 추계 학술대회 논문집 Vol.- No.-
Real-time performance is very important in realizing vision-based lane sensing system. In this paper, lane sensing algorithms are developed such that 3-D lane geometry can be reconstructed on-line from the stereo camera images. The lane geometry is modeled into two 2nd order polynomials where the parameters represent road curvature, lateral offset and heading angle, respectively. These parameters are estimated by the Kalman filter and are utilized to obtain the global lane geometry through the inverse perspective mapping. In order to accomplish the real-time ca1culation, ROIs are selected as small as possible. The algorithms are developed for 30 Frames/sec task time and are verified by the experiments implemented on a passenger car.