Locally Adaptive Bilateral Clustering for Image Denoisingand Sharpness Enhancement

The purpose of denoising is to remove the noisewhile retaining the edges and other detailed features as much aspossible. This paper, present a method for both image denoisingand sharpness enhancement. We approach the problems ofdenoising and sharpening by first adaptively segmenting theimage into clusters based on features that represent theunderlying local image structures (e.g., image details, edges,and textures). The key idea behind this approach is denoisingand sharpening according to the local image feature, so thatnoise amplification, undershoots, and overshoots can beeffectively avoided. The parameters which are used to test theperformance of the proposed method are i) Peak Signal to NoiseRatio (PSNR) and ii) Mean Absolute Error (MAE). Furtherthe results which are obtained using the proposed method interm of PSNR and MAE were compared with the earliestpublished works. The results clearly confirm the superiority ofour proposed method for sharpness enhancement and imagedenoising.