A Method of lane detection Based on a Hybrid Model in Urban Environment

Lane information is one of the main perceptive objects in the environment awareness module of intelligent vehicle. In this paper, the problem of lane detection in urban environment is considered. In order to improve the performance of lane detection in the complex environment, a novel straight-curve hybrid model is proposed. The proposed method mainly consists of coarse location and extraction of key points of lane. First, a progressive probabilistic Hough transform is used to extract the straight line. A straight-line model of lane is then developed using the color, shape, and gradient information, which can be used to locate the lane roughly. Secondly, a sliding window strategy is used to extract the key points of the curve and a curve model is constructed. Then, the straight-line model and curve model are then combined to form a novel straight-curve hybrid model. Finally, a RANSAC algorithm is used for fitting the lane. The performance of proposed method is verified by the public data set and the real data set. The results of computer simulation show that this method has better adaptability in urban environment than the existing methods.