Presegmentation of high-resolution satellite images with a multifractal reconstruction scheme based on an entropy criterium

The last generation of satellites leads to the very high-resolution images which offer a high quality of detailed information about the Earth's surface. However, the exploitation of such images becomes more complicated and less efficient as a consequence of the great heterogeneity of the objects displayed. In this paper, we address the problem of edge-preserving smoothing of high-resolution satellite images. We introduce a novel approach as a preprocessing step for feature extraction and/or image segmentation. The method we propose is related with the idea of resolution reduction and is derived from the multifractal formalism used for image compression. First, a multifractal decomposition scheme allows to extract the most singular transitions of the image. Then, an entropy-based criterium enables to consider a particular manifold composed with the most, simultaneously, relevant and singular pixels. Finally, a reconstruction scheme performed over this manifold provides an approximation of the original image. Such an approach is ideal, as it assumes that objects can be reconstructed from their boundary information, and it provides presegmented images where the main structures are preserved.