Other Free Encyclopedias » Online Encyclopedia » Encyclopedia - Featured Articles » Contributed Topics from F-J

Inpainting in Virtual Restoration of Artworks

crack digitized image color

Definition: Inpainting techniques are used to enhance the quality of the digitized artwork during virtual restoration .

Digital imaging techniques are widely used to restore, interpret and preserve collections of visual cultural heritage. However, although proper storage conditions and controlled access to the cultural heritage pieces are routinely applied, materials, such as mural, canvas, vellum, photography, and paper medium are invariably exposed to various aggressive environmental factors which lead to the deterioration of the perceived image quality. Therefore, digital solutions for archiving and popularization of heritage collections through the online access to digitized artwork virtual museums are of paramount importance.

One of the most critical issues in digitized artwork images restoration is that of crack removal and fading color enhancement. Cracks are breaks in the media, paint, or varnish of the original artwork caused mostly by aging, drying or mechanical factors. Fading colors and obscure shadows are caused mostly by environmental conditions, such as sunshine, oxidation, temperature variations, humidity, and the presence of bacteria, affect the original artwork causing. Both undesirable effects result in significant variation in the color characteristics and pigmentation of the artwork, preventing proper recognition, classification and dissemination of the corresponding digitized artwork images.

During virtual restoration the quality of the digitized artwork can be enhanced using inpainting techniques (Figure 1). For example, in the system presented in, the user manually selects a point on each crack in order to achieve the proper differentiation between damaged areas and regular pixels. Once the reference pixel is defined, an automated procedure uses the threshold values to compare the intensity of the reference pixel and its neighbors in localized image area in order to determine the crack’s mask. Note that the degree of user interaction can be reduced by utilizing a sophisticated crack detection module. After a crack is completely localized, image inpainting is used to fill-in the corresponding spatial location with image interpolated values. This step completes the restoration process. Similarly to the crack removal, a region with faded colors and obscure shadows must be first localized. Then, the user by selecting target colors from a color template and an impainting method fills-in the detected gaps and restores both intensity and color information.

User Comments

Your email address will be altered so spam harvesting bots can't read it easily.
Hide my email completely instead?

Cancel or