Robust Algorithm of Matching SAR Image to Optical Image Using Multiple Subarea

Abstract:

A novel algorithm of scene matching with spatial relation constrain was presented. The edge feature of SAR image was extracted via an improved contour detecting algorithm, which computed the gradient of pixels using wavelet packet transform. The realtime image was divided into multiple subarea and each subarea was evaluated by the gray level standard deviation and the distance of spectral histogram. The spatial relationship of the subarea was applied to correct the matching result. The multiplelayer search strategy based on lifting wavelet transform was used to improve the performance of the algorithm. The experiment result shows that the algorithm is effective and robust.

Key words:SAR scene match; edge extraction; wavelet packet transform; spectral histogram; multiple subarea