IMAGE MATCHING USING THE WINDOWED FOURIER PHASE

A theoretical framework is presented in which windowed Fourier phase ( WFP) is introduced as the primary matching primitive. Zero-crossings a nd peaks correspond to special values of the phase. The WFP is quasi-l inear and dense; and its spatial period and slope are controlled by th e scale. This framework has the following important characteristics: 1 ) matching primitives are available almost everywhere to convey dense disparity information in every channel, either coarse or fine; 2) the false-target problem is significantly mitigated; 3) the matching is ea sier, uniform, and can be performed by a network suitable for parallel computer architecture; 4) the matching is fast since very few iterati ons are needed. In fact, the WFP is so informative that the original s ignal can be uniquely determined up to a multiplicative constant by th e WFP in any channel. The use of phase as matching primitive is also s upported by some existing psychophysical and neurophysiological studie s. An implementation of the proposed theory has shown good results fro m synthesized and natural images.