CAJ | 학술논문

提出了一个基于图像区域分割和边缘检测信息融合的车道检测算法,首先在HSV颜色空间内进行基于区域的分割,得到车道边界的模板;接着将车道边界模板和道路边缘检测结果进行操作,得到车道边界点;最后对车道的边界点使用Hough变换进行处理,得到车道的方向和位置.算法将基于区域分割的信息和边缘检测的信息相互补充,提高了车道边界提取的精度.仿真实验结果表明,对于多种复杂的道路环境算法都可以鲁棒的检测出准确的车道.
제출료일개기우도상구역분할화변연검측신식융합적차도검측산법,수선재HSV안색공간내진행기우구역적분할,득도차도변계적모판;접착장차도변계모판화도로변연검측결과진행조작,득도차도변계점;최후대차도적변계점사용Hough변환진행처리,득도차도적방향화위치.산법장기우구역분할적신식화변연검측적신식상호보충,제고료차도변계제취적정도.방진실험결과표명,대우다충복잡적도로배경산법도가이로봉적검측출준학적차도.
A lane detection algorithm was proposed based on the information fusion of the region-based seg-mentation and the edge detection. Firstly, the region-based segmentation was processed in HSV color space to get the mask of lane boundary. Secondly, the mask of lane boundary and the result of lane edge detection were fused by the logic AND operator for the lane boundary. Finally, the orientation and position were obtained by Hough transform of lane boundary point. The region-based segmentation information and the edge detection in-formation mutually complete each other to detect the lane point accurately. Simulation results show that the proposed method can effectively and robustly detect the lane under different complex road conditions.