We consider in this paper the problem of image inpainting in infrared image analysis, where the objective is to reconstruct missing or deteriorated parts of an image. In this work, we develop image inpainting using quaternion representation concepts and modified exemplar-based technique. In our work, we exploit the concept of sparse representation, which takes a group of nonlocal patches with similar textures as the basic unit instead of a patch. As a result, the proposed method provides plausible restoration while propagating information of edge for the target region. Experimental inpainting results demonstrate the effectiveness of the proposed method in the task of reconstruction thermal images.
|