Disparity Measurement Based on Wavelet Phase

摘要

An algorithm for matching images of real world scenes based on multiresolution wavelet phase is presented in this paper. Disparity for a stereo pair can be calculated by solving the stereo correspondence problem in image matching. For phase matching, the difference between the complex phase at corresponding points is used to find the disparity. In this algorithm, the Hilbert transform is applied to construct a complex-valued wavelet transform kernel, and the multi-resolution analysis is combined with this kernel to get multiresolution phase of local spatial frequency components. Then the phase matching is done based on these multiresolution phase from coarse-to-fine. The feature for this algorithm is that it combines phase matching well with the way of calculating multiresolution phase.