Using a Neuro-Fuzzy Network for Impulsive Noise Suppression from Highly Distorted Images of WEB-TVs
This paper introduces a novel approach for denoising the images corrupted by Impulsive Noise (IN) by using a new nonlinear IN suppression filter, entitled t-nearest neighborhood pixels based Adaptive-Fuzzy Filter (t-AFF). The proposed filter is based on statistical impulse detection and nonlinear filtering which uses Adaptive Neuro-Fuzzy Inference System as a missed data interpolant over the t-nearest neighbor pixels of the corrupted pixels. The impulse detection is realized by using the well-known Edgington’s goodness-of-fit test which yields a decision about the impulsivity of each pixel. To demonstrate the capability of t-AFF, extensive simulations were realized revealing that the proposed filter achieves a better performance than the other filters mentioned in this paper in the cases of being effective in noise suppression and detail preservation, even when the images are highly corrupted by IN.
KeywordsImpulse Noise Impulsive Noise Noise Density Corrupted Image IEEE Signal Processing Letter
Unable to display preview. Download preview PDF.
- 3.Russo, F.: Evolutionary Neural Fuzzy Systems for Data Filtering. IEEE Instrumentation and Measurement Technology Conference 2, 826–830 (1998)Google Scholar