A new change detector in heterogeneous remote sensing imagery

Multimodal change detection in satellite images is a challenging and complex problem mainly because the local statistics of the images to be compared can be very different. In this paper, we present a novel, reliable and simple change detection operator which is first based on a imaging modality-invariant operator that detects the common specific high-frequency pattern of each structural region existing in the two heterogeneous satellite images. The resultant similarity map is then filtered out by a superpixel-based spatially adaptive filter which increases its reliability against noise. Second, in order to achieve more robustness, changes are then identified, from this similarity map, by combining the results of different automatic thresholding algorithms with a weighted spatially regularized multi-criteria decision analysis. Experimental results involving a mixture of different types of imaging modalities confirm the robustness of the proposed approach.