Image Retrieval Using Digital Image Inpainting Techniques

Abstract

Image retrieval is an inverse problem in digital image processing. In this paper, the authors deal with restoration of image using digitally image inpainting methods. In this inpainting technique, one can extract a missing an important part or can fill unwanted parts of an image, so as to recover original information without loss. They are using semi-automatic and fast image inpainting methods and compared their result in terms of image quality assessment parameters such as PSNR, MSE, as well as required timing constraints. The observational as well as statistical results show that the fast image inpainting technique is superior in all respects for any kinds of multimedia image application.