Fundus Image Enhancement using CLAHE

Authors

  • Wu Wanling Department of Electrical Engineering, Faculty of Engineering, Universiti Malaya
  • Noraisyah Mohamed Shah Department of Electrical Engineering, Faculty of Engineering, Universiti Malaya

Keywords:

Fundus retinal images; Image quality; Contrast enhancement; Histogram equalization; CLAHE.

Abstract

Fundus retinal images are crucial for ophthalmologists to diagnose diseases and monitor changes in the condition. However, due to factors such as lighting conditions, instrument effects, and individual differences, fundus images often have the drawbacks of low contrast and lack of details. To improve the quality and accuracy of images, contrast enhancement technology for fundus images has become a research hotspot. This paper proposes a new CLAHE (Contrast Limited Adaptive Histogram Equalization) method to improve the brightness and contrast of retinal images. The method improves the luminosity of fundus images by using gamma correction in the HSV color space and enhances the contrast of images by limiting contrast histogram equalization in the L*a*b* color space. Finally, the effectiveness of the method is verified through the STARE dataset. The results show that compared with the traditional CLAHE method in the RGB color space and the WAHE method, the method proposed in this paper has better improvement effects on color retinal images, and performs well in adaptability, color fidelity, local detail preservation, and algorithm implementation simplicity, making it suitable for fundus image processing under different lighting conditions. It is also easy to deploy and use in practical applications, providing reference and guidance for researchers and healthcare professionals.

Downloads

Download data is not yet available.

Downloads

Published

2025-03-13

How to Cite

Wanling, W. ., & Mohamed Shah, N. . (2025). Fundus Image Enhancement using CLAHE. New Explorations in Electrical Engineering, 1(1), 68–79. Retrieved from https://vmis.um.edu.my/index.php/NECE/article/view/59689

Most read articles by the same author(s)