PERTANIKA JOURNAL OF SCIENCE AND TECHNOLOGY

 

e-ISSN 2231-8526
ISSN 0128-7680

Home / Regular Issue / JST Vol. 26 (4) Oct. 2018 / JST-0966-2018

 

A New Genetic Algorithm Based Technique for Biomedical Image Enhancement

Khatkar, Kirti and Kumar, Dinesh

Pertanika Journal of Science & Technology, Volume 26, Issue 4, October 2018

Keywords: Bacterial foraging, denoizing, standard deviation, Fuzzy logic, genetic algorithm, Haar-wavelet, image enhancement, pre-processing

Published on: 24 Oct 2018

From a diagnostic perspective, image enhancement has diverse potential in image processing applications related to biomedical images. A hybrid algorithm obtained by combining discrete wavelet transformation with soft computing techniques is proposed for enhancing the biomedical images. This paper proposes an approach for effective visual enhancement of biomedical images. The proposed approach uses scale-invariant feature transform algorithm and principal component analysis as pre-enhancement steps, followed by the combination of DWT and the genetic algorithm to enhance the biomedical images. In GA, a new fitness function, which can efficiently reduce the noise in biomedical images while preserving the details, is proposed for the enhancement process. In order to accurately evaluate the enhanced image's quality, various metrics like peak signal to noise ratio, contrast to noise ratio, BETA coefficient, standard deviation, and mean square error have been considered. Finally, the comparison of the proposed algorithm with other soft computing techniques like Bacterial Foraging, Particle Swarm Optimization and Fuzzy Logic is carried out. The results show that the proposed technique outperformed over the other methods and provided better image quality.

ISSN 0128-7680

e-ISSN 2231-8526

Article ID

JST-0966-2018

Download Full Article PDF

Share this article

Recent Articles