Research Of Digital Image Inpainting Technique

Image inpainting is an ancient art. In renaissance of Europe, people have repaired the cracks of the work of fine art. They can resume the damaged portions of a work of fine art and keep it whole by this way. As the advent and development of computer technology, digital image inpainting was introduced into the filed of image inpainting by Bertalmio. Digital image inpainting is an important part of image processing. It is a technique, which fills in appointed damaged portions of an image based upon the available image information. This technique is used widely in many fields such as repairing damaged medical image or ancient artifact, inpainting injured photo, removing extra object in image, restoring scene of the crime and so on.Now, the most representative image inpainting at home and abroad can be divided into two kinds: the image inpainting based on partial differential equations (PDE) and the image inpainting based on texture synthesis. The image inpainting based on PDE was presented earlier. M. Bertalmio's algorithm can imitate the way of professionals' method of repair. It can be said as unprecedented in the history of the image inpainting based on PDE. Then, the total variation model and the curvature-driven diffusion model presented by Chan and Shen have often been consulted by people who needed to study the digital image inpainting. The image inpainting based on texture synthesis is also an effective method. We can get very good result when we use this way to do the big damaged blocks inpainting. The texture synthesis used in this model involves three problems which are synthesis sample,feature matching and constraint synthesis. In recent years, MRF model became the mainstream of the texture synthesis. It is because that MRF model can show the local interrelated characteristic and the stable characteristic of the texture. And the technique of patch joining has been continuously developing and improving because of its speed and quality of synthesis. In the present, the typical texture synthesis algorithms such as WL algorithm, Ashikhmin algorithm and patch-based texture synthesis algorithm are used widely by professionals.Besides, the image inpainting based on radial basis functions (RBF), which was presented by Zhou ting fang also caused wide public concern. It converts the image inpainting problem to three-dimensional implicit surface reconstruction problem firstly. With the advantages of RBF, the algorithm can repair the relatively large damaged portions more accurately and more robust.Firstly, in this paper, in a light of these kinds of image inpainting mentioned in second paragraph, we summarized the recent work and application of image inpainting. And then we studied several typical image inpainting models based on the basic knowledge of this technique; At last, we introduce the key content of this paper: a new image inpainting algorithm based on texture synthesisThis algorithm is based on MRF model. It keeps the edge feature of injured image by dealing with the priority of the boundary pixels and the weight of the pixels in neighborhood. Besides this, the texture details and structure information are fully considered. And the same time, blurring effect is avoided. This algorithm can also achieve fairly good results when do the big damaged blocks inpainting.