3D Information Acquisition from Plane Spatio-Temporal Images with Complicated Camera Motion

This paper presents a new method for estimating parameters of 3D lines from Plane Spatio-Temporal Image(PST1). A PSTI is created by using a slit to scan a 3D ST1 which is obtained by a camera mounted on a moving car. We have discussed the creation of PSTI with the cameramotion of straightline translation by using a hyperbolical slit and the 3D information acquisition before. In this paper, we give a general analysis about the formation of PSTI a t first, then propose a new method to estimate parameters of 3D lines for more complicated scanning formula and camera motion. In this method, the locus, which corresponds to a 3D line, is extracted to calculate projection vectors. And, the parameters of 3D lines are obtained from the relation between projection vectors and the parameters by recursive optimizing algorithm. The limitation of both the slit and camera motion will be decreased. At the end, some experiment results are given to reveal the effectiveness and accuracy of the method.