Application of GRNN neural network in non-texture image inpainting and restoration

摘要

Inspired by the connectivity principle of human visual perception, a new inpainting approach based on GRNN neural network is proposed in this paper. The missing regions in this new technique are determined by performing regression analysis on the image data. The missing regions are first separated and sorted according to their size. Then the algorithm proceeds with applying a GRNN network to each one in order to repair their damaged pixels. Simplicity and efficiency are the main advantages of the proposed approach. The performance of the proposed approach is evaluated in three application contexts: text removal, scratch removal, and noise removal. Where possible, we used objective measures (e.g., PSNR) to evaluate the visual quality of the inpainted i rnag:es. The results demonstrate the effectiveness of the proposed approach. (C) 2015 Elsevier B.V. All rights reserved.