Relative Performance of Histogram Equalization and Adaptive Histogram Equalization in Enhancing Low-Contrast Image

Abstract

Low-contrast images present challenges across various domains, such as computer vision, medical imaging, and digital photography, where essential details can be lost in underexposed or poorly lit conditions. Traditional contrast enhancement methods such as Histogram Equalization (HE) and Adaptive Histogram Equalization (AHE) have long been employed to enhance the visual quality of such images. HE provides a global approach to contrast enhancement by redistributing pixel intensity values to make the image histogram more uniform. At the same time, AHE improves upon this by applying contrast enhancement locally to small regions of the image, allowing for better results in images with varying lighting conditions.

This paper provides a comprehensive review of the relative performance of HE and AHE in enhancing low-contrast images, focusing on their effectiveness, computational efficiency, and suitability for various image domains. By leveraging quantitative metrics such as the Structural Similarity Index (SSIM), Peak Signal-to-Noise Ratio (PSNR), and qualitative assessments, we present a detailed comparison of these approaches. Our findings suggest that while HE and AHE are adequate for fast contrast enhancements, AHE often offers better results in complex, low-contrast images due to its localized approach. However, both methods have limitations in preserving fine details and avoiding artifacts in more challenging scenarios.