ISSN: 1007-1172

Impact Factor: 6.2

UGC-CARE APPROVED MULTIDISCIPLINARY JOURNAL

Journal of Shanghai Jiaotong University

International Peer Reviewed | Open Access | A Monthly Publishing Journal

"EFFICIENT DENOISING FOR COVID – 19 IMAGES USING WAVELET TRANSFORMS"

Authors

N. Sundaravalli, Research Scholar & DR. R. Vidyabanu, Assistant Professor
Pg & Research Department Of Computer Science, L.R.G Govt.Arts College For Women, Tirupur – 4, India.

Abstract

At this time, the entire world is facing a new contagious coronavirus disease 2019 (COVID-19) caused by novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV2). The world health organization (WHO) already declared this infectious disease as a worldwide pandemic. During this period, the medical X-ray image plays an important role to diagnose COVID- 19 patients effectively. This current research work takes the dataset of COVID-19 affected chest X- ray and proposes a new method of image de-noising based on using median filter (MF) in the wavelet domain. In this contemporary exploration, various types of wavelet transform filters are used in conjunction with median filter, in experimenting with the proposed approach in order to obtain better results for image de-noising process and, consequently to select the best-suited filter. Wavelet transforms mathematical tools working on the frequencies of sub-bands split from an image in different scales. According to this experimental work, the proposed Dual-Tree Complex Wavelet Transform (DT_CWT) method presents better results than using only wavelet transform methods. Here, by working on real time images and later adding noise (salt & pepper, Gaussian) to images and then calculate and comparing the Peak Signal to Noise Ratio (PSNR), Mean Square Error (MSE), Structural Similarity Index (SSIM) value for different images. The present work aims in categorizing the x-ray images as COVID-19 infected and not infected images to facilitates the effective diagnosis of the disease and thus can be used to inform future Literature-Based Discovery endeavors.