Steerable filter based multiscale registration method for JERS-1 SAR and ASTER images

It is a challenging problem to automatically register the SAR and optical satellite images in remote sensing applications. In this paper, a novel intensity based multiscale registration method using steerable Simoncelli filters is proposed to register JERS-1 SAR and ASTER images. The mutual information is used as the similarity measure, and a hybrid search technique is used to do the parameter optimization. The experimental results showed that the proposed registration scheme is competent for JERS-1 SAR and ASTER images, though the robustness of this scheme should be verified further more.