Image denoising based on statistical jump regression analysis and local segmentation using Normalized Cuts

The Edge-Preserving Surface Estimation based on statistical jump regression analysis is a powerful approach for image denoising. However, it requires an accessorial corner-preserving technique in which a corner threshold needs to be tuned. In this paper, we suggest a novel procedure based on local segmentation using Normalized Cuts which can well preserve the edges and corners at the same time without using the corner-preserving technique. Extensive experiments show that the proposed approach outperforms the state-of-the-art existing approaches.