Road environment information acquisition for vehicles based on optical flow analysis
Qijie Zhao · Journal of Hefei University of Technology · 2009
The traditional headlight system has some hidden problems of safety,but the adaptive front-lighting system(AFS) can eliminate these problems and reduce the rate of traffic accidents effectively.Achieving vehicle running surrounding feature information accurately is the precondition of AFS implementation.In this paper,the image sequence is obtained from the video data of driver's vision when the vehicle is running at night.A kind of image region segmentation method is proposed and the image motion estimation method based on the optical flow model is used to calculate the optical flow velocity of each pixel in four regions,then a rule of judging vehicle running surrounding features is proposed according to the velocity to obtain road environment information when the vehicle is running at night.The effectiveness of the method is proved.