Moving Target Detection and Tracking in Vision Navigation of the Vehicle on the Road

Motion this paper researches on moving target detection and tracking in automatic visual road vehicle guidance system by processing and analysing a long sequence images of real road which are gathered on a vehicle driving outside. The core of our algorithm is to detect a target by finding the horizontal and vertical edges, and to verify whether a target exists or not by computing the local rear symmetry of the target in the image. And building the state model that can represent the relative motion between the target and the camera, the perspective model that can represent the features of th the target, then estimate the 3D-motion parameters by generalized nonlinear Kalman filter.