화상처리 기법을 이용한 도로의 백색선 인식에 관한 연구
- Alternative Title
- A Study on Real-Time Recognition of Road Lane for Lateral Control of Autonomous Vehicle
- Abstract
- 본 논문에서는 차량의 횡방향 제어알고리즘을 개발하기 위한 선행 단계로 CCD카메라에서 획득한 디지털 영상정보를 이용하여 백색 차선을 인식하는 알고리즘을 제시한다. 차선은 역원근 처리와 폭 정보를 이용한 경계선처리를 통해 원 화상을 차선의 명도가 강화된 형태로 변형시킨 다음, Metropolis 표준법을 이용한 Simulated Anealing 알고리즘을 사용하여 주어진 화상에서 차선일 확률이 가장 높은 차선모델을 선택하는 방법으로 인식한다. 이 차선인식 알고리즘은 PC급 컴퓨터성능의 비젼시스템에서 실시간으로 차선을 인식하며 특히 그림자가 있거나 불연속적인 경우 등의 다양한 조건에서도 차선을 인식할 수 있어 실차 적용의 가능성이 확인되었다.
This paper introduces a lane-recognition algorithm for detecting lane markers on digital images acquired from CCD camera and a lateral-control algorithm for keeping a reference path following.
The first-stage of lane recognition includes grey level digitization, the edge detection of white lane markers and inverse perspective remapping.
The next-stage of lane recognition is to find parameters of quadratic polynomial assumed as a road model with Maximum probability estimation. The estimation method is implemented by Metropolis Algorithm on PC-based computer system. The rate of recognition is about 33 frames/second. Also, it has been shown to recognize driving lane robustly, even in the presence of shadow, discontinuity of lane marker and reflection of road surface.
This paper introduces a lane-recognition algorithm for detecting lane markers on digital images acquired from CCD camera and a lateral-control algorithm for keeping a reference path following.
The first-stage of lane recognition includes grey level digitization, the edge detection of white lane markers and inverse perspective remapping.
The next-stage of lane recognition is to find parameters of quadratic polynomial assumed as a road model with Maximum probability estimation. The estimation method is implemented by Metropolis Algorithm on PC-based computer system. The rate of recognition is about 33 frames/second. Also, it has been shown to recognize driving lane robustly, even in the presence of shadow, discontinuity of lane marker and reflection of road surface.
- Author(s)
- 박성태; 정종화
- Issued Date
- 2002
- Type
- Research Laboratory
- URI
- https://oak.ulsan.ac.kr/handle/2021.oak/4037
http://ulsan.dcollection.net/jsp/common/DcLoOrgPer.jsp?sItemId=000002025093
- Authorize & License
-
- Files in This Item:
-
Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.