Impulse Noise Detection and Removal Method Based on Modified Weighted Median

Impulse Noise Detection and Removal Method Based on Modified Weighted Median

Ashpreet (Department of Computer Engineering, NIT, Kurukshetra, Haryana, India) and Mantosh Biswas (Department of Computer Engineering, NIT, Kurukshetra, Haryana, India)
Copyright: © 2020 |Pages: 16
DOI: 10.4018/IJSI.2020040103

Abstract

Impulse noise generally occurs because of bit errors in progression of image acquisition and transmission. It is well known that median filtering method is an impulse noise removal method. Lots of modified median filters have been proposed in the last decades to improve the methods for noise suppression and detail preservation, which have their own deficiencies while identifying and restoring noise pixels. In this article, after deeply analyzing the reasons, such as decreased noise detection and noise removal accuracy that forms the basis of the deficiencies, this article proposes a modified weighted median filter method for color images corrupted by salt-and-pepper noise. In this method, a pixel is classified into either “noise free pixel” or “noise pixel” by checking the center pixel in the current filtering window with the extreme values (0 or 255) for an 8-bit image using noise detection step. Directional differences and the number of “good” pixels in the current filtering window modify the detected noise pixels. Simulation effects on considered test images reveal the proposed method to be improved over state-of-the-art de-noising methods in terms of PSNR and SSIM with pictorial comparative analysis.
Article Preview
Top

1. Introduction

The sensor malfunctions at some stage in image acquisition, hardware failures in the course of storage of image, interventions in the course of transmission of image, and artifacts throughout compression and processing of image frequently degrades the images by impulse noise (Plataniotis & Venetsanopoulos). Salt-and-pepper (SAP) noise is communal in natural images, with maximum and minimum intensities. In an 8-bit gray level image, this shows up as white and black points that look like “salt” and “pepper,” individually. SAP noise is created for the most part during the procedure of image capture and storage, because of false localities in memory and broken image sensors.

In image enhancement, de-noising of images is required for several image processing applications (Pitas & Venetsanopoulos, 1992; Tukey, 1977; Roy & Laskar, 2017; Oussama et al., 2018; Aboali et al., 2018). There are three forms of strategies used for reducing impulse noise in color images: channel-wise strategies, vector filtering strategies and various other strategies such as PDE based (Kim, 2006), wavelet domain based (Brook 2015), soft computing based (Schulte et al., 2006; Nair & Raju, 2012) and machine learning based (Roy & Laskar, 2016). Channel-wise strategies put on algorithms used for grayscale images to de-noise individual color channels (Koschan & Abidi, 2008). The classical vector filters are the median type filters (Chanu & Singh, 2016) with vector median filter (VMF) (Astola et al., 1990), basic vector directional filter (BVDF) (Trahanias & Venetsanopoulos, 1993), and directional distance filter (DDF) (Karakos & Trahanias, 1997). These filters work out median vectors which depend on various gathered vector distance checking principles. However, those filters result in smoothing and destroying the structures of edges & info, since it creates all pixels further irrespective of whether the pixels are noisy or not. The naivest resolution concerning this problem is the recognition of each pixel as noisy or not after which restoration of most effective the noisy pixels. This thoughtful method is well known as “Switching Vector Median Filter (SVMF)” (Smolka & Chydzinski, 2005; Celebi & Aslandogan, 2008; Celebi et al., 2010; Matsuoka et al., 2016).

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 9: 4 Issues (2021): Forthcoming, Available for Pre-Order
Volume 8: 4 Issues (2020)
Volume 7: 4 Issues (2019)
Volume 6: 4 Issues (2018)
Volume 5: 4 Issues (2017)
Volume 4: 4 Issues (2016)
Volume 3: 4 Issues (2015)
Volume 2: 4 Issues (2014)
Volume 1: 4 Issues (2013)
View Complete Journal Contents Listing