Open AccessProbabilistic Decision Based Average Trimmed Filter for the Removal of High-Density Salt and Pepper Noise
Author(s) -
Amit Prakash Sen,
Nirmal Kumar Rout
Publication year - 2020
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.e6739.018520
Subject(s) - salt and pepper noise , median filter , noise (video) , computer science , filter (signal processing) , artificial intelligence , probabilistic logic , noise spectral density , mathematics , computer vision , image processing , image (mathematics) , telecommunications , noise figure , amplifier , bandwidth (computing)
The paper focuses on the evacuation of salt and pepper noise from a contaminated image. A probabilistic decision based average trimmed filter (PDBATF) is proposed for both high and low noise density. The proposed algorithm addresses the issue related to even number of noise-free pixel in trimmed median filter for the calculation of processing pixel. The proposed average trimmed filter is incorporated for low noise density while the proposed patch else average trimmed filter is applied for high noise density. Finally, they are combined together to develop the proposed PDBATF. The proposed algorithm show an excellent noise removal capability compared to the recently developed algorithms in terms of peak signal to noise ratio, image enhancement factor, mean absolute error and execution time. It works very efficiently in de-noising contaminated medical images such as chest-x-ray and malaria-blood-smear.