2018/11/01 by Mohammad Tariqul Islam, Dipayan Saha, Islam, Mohammad Tariqul +7
Computer Science · Engineering · #Advanced Image Fusion Techniques #FOS: Electrical engineering #Image Processing Techniques and Applications #Image and Signal Denoising Methods #Image and Video Processing (eess.IV) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1811.00244
openalex publication_date 2018/11/01 · openalex created_date 2022/08/02 · openalex updated_date 2026/07/28
Reduction of mixed noise is an ill posed problem for the occurrence of\ncontrasting distributions of noise in the image. The mixed noise that is\nusually encountered is the simultaneous presence of additive white Gaussian\nnoise (AWGN) and impulse noise (IN). A standard approach to denoise an image\nwith such corruption is to apply a rank order filter (ROF) followed by an\nefficient linear filter to remove the residual noise. However, ROF cannot\ncompletely remove the heavy tail of the noise distribution originating from the\nIN and thus the denoising performance can be suboptimal. In this paper, we\npresent a variational step to remove the heavy tail of the noise distribution.\nThrough experiments, it is shown that this approach can significantly improve\nthe denoising performance of mixed AWGN-IN using well-established methods.\n