Digital Image Repair Technology

Digital image inpainting is a technique of repairing a partially damaged or missing image in an undetectable way. It fills the missing part of an image to make it look natural after repairing by employing information of the undamaged part according to some rules. Digital image inpainting derives from restoration of artworks, and has been applied to repair ancient paintings. With the wide spread of multimedia, this technique becomes an important task in a variety of applications, such as object removal, super resolution, image compression and video error concealment.In theory, the inpainting problem is clearly ill-posed, because there is no sufficient information available to guarantee the existence and uniqueness of a solution. Based on studies of visual psychology and perception, researchers proposed various assumptions and constraints to solve the problem. Therefore, image inpainting is essentially a process of learning, understanding and realization of visual perception, and it plays an important role for research on computer vision.Based on the above considerations, we study digital image inpainting techniques in this dissertation. After a detailed analysis of existing methods, we proposed some algorithms to improve the performance of inpainting. The main contribution of this dissertation can be summarized as follows.1) We studied the inpainting problem from a view of visual psychology, and summarized some guidelines for inpainting by analyzing existing experimental results of visual perception and fill-in of retinal scotomas. Based on these guidelines, we conducted a detailed analysis and discussion of existing inpainting methods.2) We proposed an adaptive approach for structure inpainting based on the total variation model. In the previous inpainting method, there is a contradiction between the speed of information propagation and the sharpness of repaired image edges. Our proposed approach provides a solution for this problem and has a better visual performance. Moreover, the computation time is reduced considerably.3) We proposed an inpainting method based on the Poisson equation. In this method, the image to be repaired is decomposed into the structure image and the texture image, which are processed respectively according to their image characteristics. The texture image is repaired by the exemplar-based method; for the structure image, the Laplacian operator is employed to enhance the structure information, and the Laplacian image is inpainted by the exemplar-based algorithm, followed by a reconstruction based on the Poisson equation. By this means, the repaired image appears more natural. An incomplete Cholesky conjugate gradient method is employed to. solve the Poisson equation, which makes computation of the solution more stable and faster.