Vehicle position and speed estimation method based on binocular sequence images

The invention discloses a vehicle position and speed estimation method based on a binocular sequence image, and the method comprises the following steps: S1, obtaining a depth map and a point cloud map through a ZED binocular camera; S2, achieving background subtraction through a KNN algorithm, and identifying a moving target and a static background environment in the sequence image; S3, detectingedge points of the moving target, drawing a rectangular identification box located on the moving target, and tracking the moving target in real time; S4, removing a rectangular identification frame appearing on the non-moving object caused by error factors such as light and shadow; S5, calculating pixel coordinates of a central point of an effective rectangular identification box locked on the moving target; S6, obtaining (X, Y, Z) three-dimensional space coordinates of the points; and S7, estimating the position and the speed of the next moment according to the three-dimensional space coordinates of the previous fra