Unstructured-lane detection based on trapezoidal model and SVM

This paper presents a method of unstructured lane detection based on trapezoidal model proposed by H.Jeong,et al and SVM(Support Vector Machine).The frames extracted from the video are pretreated by PCNN(Pulse Coupled Neural Network), and then processed by Kalman filter and EM(Expectation Maximization) algorithm.Using SVM get the result of the lane detection and using morphological filter get the final detecting result.Because this method uses SVM,which has a better classified performance than BP(Back Propagation)neural network,it obtains a better detection result than that using BP neural network(H.Jeong, et al).Furthermore,this method uses PCNN to process the frames to remove the shadow in the road,reducing the effect of illumination variations.Experimental results show that this method can receive better lane detecting results than the trapezoidal model and BP neural network proposed by H.Jeong,et al.