Integration of Variational Optical Flow and Surface Splines for Dense Correspondence Estimation of Remotely Sensed Images

Abstract

A novel composite approach through integration of variational optical flow and surface splines is presented to obtain sub-pixel accurate dense disparity map for remotely sensed stereo image pair. It is well known that, surface splines handle geometric distortion very well. The performance of surface splines for dense correspondence can be significantly improved by the reliable control points, but can also be degraded by false matches significantly. To maximize the number of reliable control points, optical flow for each pixel is estimated by global energy minimization using variational method. Here, the control points are selected using two validation techniques: left-right consistency check and epipolar geometry check using fundamental matrix. In this paper, the effectiveness of the proposed approach is demonstrated on remotely sensed stereo image pair.