On the Generalization of Digital Total Variation Filter

Image denoising, as a low-level step of image processing, is perhaps the oldest and still active research topic. A general formulation based on the variation of digitalized energy to denoise image is proposed in this paper. This method is different from classical variational method employed in image processing. For a digitalized energy functional, we first compute the variation, then design algorithms leading to digital filters. Numerical experiments and comparative examples are thus carried out to verify the effectiveness of the proposed method, which is efficient, adaptive and easily implemented. Higher quality images can be obtained with characteristic singular features preserved.