A Method of Image Enlargement Based on Constrained Pixels

We propose an image enlargement method preserving perceptual sharpness, which is achieved by augmenting a low resolution image with constrained pixels from a given image itself. In this approach, we decompose the whole image into interior image and boundary image. We firstly enlarge the interior image by using 10 constrained pixels located in 5times5 neighborhood. Next, boundary image is enlarged by using 6 constrained pixels located in 3times3 neighborhood. As a distance measure for looking for the optimal pixels, we employ the Euclidean distance which is a local distance measure incorporating pixel correlation. The effective-ness of the proposed method is verified by some image enlargement experiments. Consequently, the experimental results show that the performance of the proposed method is superior to other conventional image enlargement methods objectively and subjectively based on the image quality evaluation system.