Abstract: In order to resolve the problem of lane detection in Micro-traffic under complex illumination,a lane detection method was proposed for the Micro-traffic under complex illumination．Firstly,the lane images for different illumination are classified by using Naive Bayes．Then,the classified lane images are proceed by using the corresponding image processing method.Finally,improved Otsu and revised Probabilistic Hough Transform are introduced to detect lanes．Simulation experiment on the different lane images under different illumination shows that the lane detection success rate in Micro-traffic is up to 95.5% and the method possesses strong robustness and anti-interference．

Key words: Intelligent transportation system(ITS),Naive bayes,Otsu,PPHT