-Regularized Deconvolution Network for the Representation and Restoration of Optical Remote Sensing Images

Optical remote sensing images of land cover are composed of many natural and man-made objects and thus exhibit rich image features ranging from low to high levels. Extracting such a wide range of features to represent the optical remote sensing images, beyond edge primitives, is a long-standing goal in the remote sensing and vision research community. The recently proposed deconvolution network (DN) can effectively learn and capture features in a variety of forms: low-level edges, midlevel edge junctions, high-level object parts, and complete objects. The approach is based on the convolutional decomposition of images under an L₁ sparsity constraint. Unfortunately, the L₁ regularizer cannot enforce further sparsity, hence limiting the practical efficacy of the DN in optical remote sensing representation and processing. In this paper, we extend the DN by incorporating the L_1/2 sparsity constraint, which we name the L_1/2-DN. The L_1/2 regularizer not only induces sparsity but is also a better choice among L_q, (0<;q<;1) regularizers. Furthermore, the L_1/2-DN algorithm is more efficient, provides a sparser representation, and results in more accurate recovery than the DN. We illustrate the utility of our method on a wide range of optical remote sensing images and compare our results to those yielded by other state-of-the-art methods.