As non-structural road becoming an important part of the intelligent vehicle driving environment, such as rural road, desert terrain and other different complicated field environments, machine vision navigation seems more difficult and surely to be our research priorities. According to the complex road environment of autonomous intelligent vehicles, a new road detection and recognition algorithm is proposed. This algorithm uses the genetic algorithm for image segmentation, and the mathematical morphology revise is used for the divided image. Then the canny operator is used for the road edge detection, and finally the path boundary information is obtained accurately and clearly. The simulation results show that this method can effectively divide the segment of the roads and can adapt to the various road conditions significantly.%随着将智能车辆适用环境拓展到乡村道路、荒漠、越野环境等非结构道路,机器视觉导航的难度进一步加大,成为当今人们研究的重点.文章针对智能车辆行驶的复杂道路环境提出一种新的道路检测与识别的算法.该算法采用了遗传算法对道路图像进行分割,并在此基础上对已分割图像进行数学形态学处理,然后利用canny算子进行道路边缘检测,最终得到准确清晰的道路边界信息.实验结果表明,该方法可以有效地分割出道路区域,对多种复杂路况具有良好的适应性.
展开▼
