This paper presents an approach to detect multiple lane and vehicles. Instead of assuming that the processes of lane and vehicle detection are independently, we integrate these two processes in a mutually supporting way to achieve more accurate results. In lane boundary detection, the features of lane boundary often affect by the edge and color of the vehicle. Furthermore, the results of vehicle detection could be non-robust if there are some non-vehicle objects that have similar features to vehicle. Here, we use the distance of the position between central position of lane boundary and vehicle position from hypotheses to filter out the non-vehicle object.And we use the similarity of the lane boundaries direction and the moving direction from hypotheses to get the optimal lane solution. By applying iterative optimization algorithm, we can achieve sub-optimal solution of lane and vehicle detection and the experimental results shows that the error rate is successfully reduced from 32.6% to 2.7%.

Understanding the road scene structure is essential and important for perceiving the driving situation in intelligent transportation systems (ITS). In this paper, we aim at analyzing the road scene structure by classifying the pixels to three different types, including road surface, lane markings, and nonroad objects. Instead of detecting these three objects separately in traditional approaches, we integrate different ad hoc methods under the conditional random field framework. Three feature functions based on three cues including smoothness, color and lane marking segmentation, are used for pixel classification. Besides, an optimization algorithm using graph cuts is applied to find the solutions efficiently. Experiments on the data sets demonstrate high classification accuracy on objects in the road scene.