Digital Image Inpainting Techniques for Cultural Heritage Preservation and Restoration

Physical artefacts or ancient artworks including manuscripts and photographs at heritage sites are likely to have significantly damaged or broken-down regions. Therefore, digital image restoration is utilised in cultural heritage (CH) preservation and restoration by taking digital photographs of the historical artefacts. Then, numerous image processing techniques including image denoising, image decomposition, image deblurring, image inpainting, etc. can be applied as a post-processing step to improve the quality of these photographs. Of these techniques, image inpainting plays a crucial role in filling missing parts and recovering damaged regions of digital images. This chapter presents several digital image inpainting techniques that can be used to perform inpainting for CH. Also, this chapter presents a new two-stage inpainting method for cultural heritage digital image using Delaunay triangulation-based interpolation and exemplar-based inpainting method. This method, first, assigns colours to the unknown pixels, i.e. the selected region to be inpainted, from the nearest known pixels of the images using Delaunay triangulation-based interpolation. Then, it employs an exemplar-based algorithm to perform the inpainting on the interpolated image. The experimental results showed that our proposed method can produce better results than the stand-alone exemplar-based inpainting method. Moreover, examples of inpainting tasks showed the potential of our proposed method in producing high-quality inpainting results and, at the same time, overcome the limitations of exemplar-based inpainting.