Image Zooming Based on Cartoon and Texture Decomposition
Dong Huan Jiang, Guang Bao Xu · Advanced materials research · 2012
A new algorithm for image zooming based on cartoon and texture decomposition is presented in this paper. The basic idea is to first decompose the image into cartoon and texture, and then zoom each part separately with different image zooming algorithms. Finally, the zoomed images will be synthesized into one image. The zoomed parts of the image are found by minimizing the different variational functional in the wavelet domain which use the Besov norm to measure the regularity of the parts. Unlike the traditional image zooming by interpolation, the variation model and image cartoon-texture decomposition is incorporated in the zooming algorithm. Experimental results have verified the validity of the new algorithm.