Research on Road Recognition Algorithm Based on Structure Environment for ITS

Median value filter is used to eliminate or reduce the noise caused by the outside environment. The edge information of road image is enhanced by Sobel operator. And then based on the Otsu method, an effective algorithm is proposed to binary-code the lane image in order to segment and identify road edge easily. Finally, Hough transformation is used to extract the feature of lane and detect the lane edge. Thus road edge is detected and area of interest (AOI) is established. Based on the position correlation in the video images, the Kalman filter is introduced to predict the position of AOI to reduce the searching area. This algorithm integrates the Kalman filter and precognition of road position, and improves the rapidity and accuracy of road detection.