Open AccessImpulse Noise Removal in Mammograms using Bi-Dimensional Empirical Mode Decomposition and Fast Adaptive Bilateral Filter
Author(s) -
Sannasi Chakravarthy,
Harikumar Rajaguru
Publication year - 2019
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.b1685.078219
Subject(s) - impulse noise , computer science , salt and pepper noise , bilateral filter , pixel , artificial intelligence , median filter , hilbert–huang transform , noise (video) , computer vision , adaptive filter , filter (signal processing) , impulse (physics) , noise reduction , pattern recognition (psychology) , algorithm , image processing , image (mathematics) , physics , quantum mechanics
The work aims to detect and correct the noisy mammogram images corrupted by impulse noise. This is achieved in two phases – identification of noise-affected pixels and renovation of those pixels in an image. The pixels which are disturbed by impulse noise are identified by Bi-dimensional Empirical Mode Decomposition (Bid-EMD). The restoration of these pixels and noise removal are done by fast adaptive bilateral filter (fABF). The proposed work for impulse noise removal is examined on digital mammogram images of Digital Database for Screening Mammography (DDSM) database. The proposed approach is compared with other existing state-of-the-art schemes using peak signal to noise ratio (PSNR) and image enhancement factor (IEF) performance measures. The study of performance of the proposed scheme provides enhanced outcome than the other algorithms used for impulse noise removal.