Research Of Image Inpainting Based On Exemplar Patch

The damaged area must be inpainted in order to restore the damaged image and meet people's demand of high-quality printing. In this paper, a kind of technlogy on image inpainting based on exemplar patch is provided, which combined the technology of structure inpainting and the technology of texture synthesis. The fruits of the research on the technology can be applied to such fields as archaeological fields, art, military , advertisement. It can improve the quality of image inpainting and improve people's visual effect, and it especially has very important significance on protecting historical relic and restoring historical materials.On the foundation of the basic principle of image processing, an algorithm based on exemplar patch on image inpainting is proposed by combing the theory of isophote-driven and the principle of the best sample matching. And the research on inpainting damaged image automatically on computer is carried on by using the algorithm. The core of the algorithm is an isophote-driven image-sampling process, which mainly consists of four steps: Obtaining the patch with the highest priority in the fill front, finding the best exemplar patch in the exemplar region , copying the relevant data from the best exemplar patch and updating the confidence values of pixels which have been filled just now. This algorithm prefect the mechanism of image inpainting, has important meanings on remedying the defect of image inpainting technology , and can offer a kind of new thinking on developing the inpainting technology.Experimental result indicates that the method can be capable of filling both texture and structure of the damaged areas of image simultaneously. In inpainting damaged linear structure of texture image, its advantage can be stressed especially. In addition, there are obvious results when it is applied to object removal. Compared with the existent inpainting algorithms, it has obvious virtue at the color difference of the inpainted image, the PSNR value of the inpainted image and the inpainting speed.