Development of an Improved Adaptive Switching Approach for Impulse Noise Reduction in Magnetic Resonance (MR) Images
Loading...
Date
2019-06-11
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
MPUT, UDAIPUR
Abstract
Extensive use of digital imaging in medicine today, the quality of medical
images has becomes an important issue. Medical images such as MRI have been
exploited for diagnosis as well as for more pathological truthful changes. It suffers
from number of shortcomings and this includes: ambient noise from environment,
acquisition noise from equipment, presence of other organs, background tissue and
anatomical influence such as body fat, and breathing motion. Various types of noise
generated limits the effectiveness of medical image diagnosis. Among all the noise
present in digital images, impulse noise is a major type of noise, which is generated
during medical operations such as angiography, CT- Scan, MRI. It could degrade the
image quality and cause some loss of information details. It results in misleading the
experts in the process of further diagnosis. Therefore, noise reduction is very
important.
This thesis discusses an improved adaptive switching algorithm for noise
removal in magnetic resonance (MR) images for the restoration of gray scale and
color images that are corrupted by the random valued impulse noise. It comprises of
Discrete Cosine Transform (DCT) and Median Filter in collaboration with Particle
Swarm Optimization (PSO). The filtration process is applied to the noisy pixels only
and the noisy pixel is optimized using swarm intelligence in an efficient manner. The
proposed algorithm is tested on the normal as well as medical brain images by varying
noise density from 10% to 60%. The research work is implemented in the MATLABĀ®
2018b environment.
The proposed approach outperforms in terms of Mean Square Error (MSE),
Peak Signal to Noise Ratio (PSNR) and Structural Similarity Index (SSIM). The
research will help to improve detection and prognostic capabilities in the medical
applications.
Description
Development of an Improved Adaptive Switching Approach for Impulse Noise Reduction in Magnetic Resonance (MR) Images
Keywords
Citation
Shrimali A. And Agrawal N.