Hybrid UM-LT-AHE Technique for Contrast Enhancement of Medical Images

This paper is concerned with combinations of un-sharp masking, logarithmic transformation, and adaptive histogram equalization techniques to arrive at a hybrid method for enhancement of different types of medical image's contrast. Motivation behind the hybridization is the need to have a contr...

Full description

Saved in:
Bibliographic Details
Main Authors: Abolaji Okikiade Ilori, Kamoli Akinwale Amusa, Olumayowa Ayodeji Idowu
Format: Article
Language:English
Published: Institute of Technology and Education Galileo da Amazônia 2024-12-01
Series:ITEGAM-JETIA
Online Access:https://itegam-jetia.org/journal/index.php/jetia/article/view/1280
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper is concerned with combinations of un-sharp masking, logarithmic transformation, and adaptive histogram equalization techniques to arrive at a hybrid method for enhancement of different types of medical image's contrast. Motivation behind the hybridization is the need to have a contrast enhancement method that is not application-specific and that can be deployed to several medical image enhancements. Four different types of medical images: X-ray, ultrasound, magnetic resonance, and computer tomographic images are utilized in the evaluations of the proposed hybrid contrast enhancement method. As performance metrics, absolute mean brightness error, mean square error, peak signal to noise ratio, and entropy are used. Comparative results, both qualitative and quantitative, were conducted at the end of the research, and the proposed method outperformed the other three (CLAHE, fuzzy-based, and wavelet transform-based) related selected methods in the field that used the same dataset in terms of testing accuracy. The enhancement quality of the proposed method was found to be satisfactory and can be used for any time of medical image; thus, the proposed hybrid technique produces better enhanced medical images from different medical image inputs.
ISSN:2447-0228