Image Denoising Using Novel Social Grouping Optimization Algorithm with Transform Domain Technique

Image Denoising Using Novel Social Grouping Optimization Algorithm with Transform Domain Technique

B V D S Sekhar (Andhra University, Visakhapatnam, India), P V G D Prasad Reddy (Andhra University, Visakhapatnam, India), S Venkataramana (Andhra University, Visakhapatnam, India), Vedula V S S S Chakravarthy (Raghu Institute of Technology, Modavalasa, India) and P Satish Rama Chowdary (Raghu Institute of Technology, Modavalasa, India)
Copyright: © 2019 |Pages: 13
DOI: 10.4018/IJNCR.2019100103

Abstract

In recent days, image communication has evolved in many fields like medicine, entertainment, gaming, mail, etc. Thus, it is an immediate need to denoise the received image because noise that is added in the channel during communication alters or deteriorates information contained in the image. Any image processing techniques concerned with image denoising or image noise removal has to be started with the spatial domain and end with the transform domain. A lot of research was carried out in the spatial domain by modifying the performance of different image filters such as mean filters, median filters, Laplacian filters, etc. Recently much research was carried out in Transform techniques under the transform domain, with evolutionary computing tools (ECT). ECT has proven to be dominant when compared with traditional denoising techniques in combination with wavelets in the transform domain. In this article, the authors applied a novel ECT such as SGOA on the denoising problem for denoising monochrome as well as color images and performance for denoising was evaluated using several image quality metrics.
Article Preview
Top

Introduction

Noise introduced into the image during communication disrupts pixel data embedded in the image. Researchers proposed many traditional techniques for denoising of an image and restoring the image information quality. Recently a number of intuitive techniques were developed which are metaheuristics, and are inspired by nature were proposed to take advantage and delimit limitations of conventional techniques. Such Mata Heuristic techniques proved more efficient employing on image compression, robust-watermarking, image denoising and other inter disciplinary applications (Luisier et al., 2007; Buades, et al., 2002; Buades, et al., 2005; Azzabou et al., 2007; Mohammad Reza Bonyadi et al., 2017; Andries et al., 2007; Kennedy et al., 20011; Yuhui Shi et al., 2001; Yuhui Shi et al., 1998; Eberhart et al., 1995; Eberhart et al., 2000).

A lot of research has carried out in spatial domain by modifying the performance of different image filters such as mean filters, median filters (Bhandari et al., 2016; Ashish et al., 2015; Sekhar et al., 2015), Laplacian filters, etc. Recently much research is carried out in Transform techniques under transform domain, with evolutionary computing tools (ECT). ECT has proved to be dominant when compared with traditional denoising techniques in combination with Wavelets in transform domain. (Sekhar et al., 2017).

In this article authors made an effort towards applying novel evolutionary computing tool like Social Grouping Optimization Algorithm (SGOA) in combination with the advantages of wavelet theory for image denoising by making intuitive decision on the wavelet parameters using evolutionary computing tool (SGOA) is demonstrated considering an image deteriorated with Gaussian noise. The novel denoising algorithm SGOA is discussed principally and its implementation. Later authors combined SGOA with Discrete wavelet transform techniques (DWT) which was implemented in this article. The Figure 1 shows a methodology of proposed novel method. An Objective function or Cost function was derived and defined to evaluate the performance of proposed methodology by analyzing the obtained result. Authors have proposed Flower Pollination algorithm (FPA) and evaluated its performance on denoising of natural images (Sekhar et al., 2018) using different image quality metrics.

The organization of article is done as, detailed illustration of the SGOA is specified in Section 2 and Section 3 illustrates the implementation of proposed novel method. Section 4 deals with analysis and results. Finally, Section 5 focuses on conclusion of the work and possible further enhancement of work. Authors have used MATLAB platform to carry out experiments. The image dataset is standard images from MATLAB tool under image processing tool box.

Figure 1.

General Flow of Proposed Technique denoising

IJNCR.2019100103.f01

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 9: 4 Issues (2020): 1 Released, 3 Forthcoming
Volume 8: 4 Issues (2019)
Volume 7: 4 Issues (2018)
Volume 6: 2 Issues (2017)
Volume 5: 4 Issues (2015)
Volume 4: 4 Issues (2014)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing