基于边界的车道标识线识别和跟踪方法研究

在道路图像中，为了得到较理想的车道的标识边缘，该文采用基于LOG算子边缘增强的方法得到梯度图像，利用SUSAN算子对该梯度图像进行分割，实现了车道标识线边界的提取。在分割后的梯度图像中，利用Hough变换提取出车道标识线参数，完成了对道路图像中的车道标识线的识别，采用建立梯形的感兴趣区域的办法，实现了对车道标识线的实时跟踪。试验结果表明该方法具有较强的鲁棒性和较好的实时性。

Research on Linear Lane Mark Identification and Track Method Based on Edge

In order to get an ideal lane marks' edge in a road image, an enhance edge method based on LOG operator is used to acquire a grads image, then the grads image is divided up by SUSAN operator, the lane mark edges is picked-up. In the divided up image, it is realized to identify lane marks because lane mark line parameters can be acquired by Hough transform, the lane mark real-time track is performed by establishing trapezoid areas. Experimental results show that the proposed method is characterized by reliability, good robustness and real-timeness.