Superrésolution aveugle d'images par la méthode des sous-espaces

In this work, we present a method for solving the blind superresolution problem. A set of images, acquired from the same scene through various unknown filters, is used to reconstruct a high resolution image. We use the subspace method to estimate the unknown filters. It consists in the study of the second order statistics of the dierent observations in order to recover the filters. We show how to specify this method in the context of superresolution.