Low-Contrast Color Image Enhancement with Improvised Histogram Equalization based on Spatial Contextual Information

Abstract

Enhancement of Color images plays a vital role in applications like medical imaging, remote sensing, and surveillance by improving visual quality by adjusting brightness, contrast, and sharpness. However, traditional techniques often suffer from noise amplification, over-enhancement, and loss of detail, particularly in low-exposure images with poor visibility. To address these challenges, this paper introduces Exposure-Based Sub Histogram Equalization with Spatial Contextual Information (RS-ESIHE-E), a novel approach that replaces conventional histograms with an Energy Curve derived from spatial contextual information, enabling more precise contrast enhancement while preserving critical details. The proposed methodology combines the advanced technique, Recursive Symmetric Exposure-based Sub-image Histogram Equalization (RS-ESIHE), and the concept of Spatial Contextual Information, in which the histogram is replaced by an energy curve, which independently enhances different exposure regions. This method was tested on low-contrast color images and compared against established techniques. Performance was evaluated using metrics such as AMBE (brightness preservation), PSNR & MSE, Entropy (information richness), SSIM & FSIM (perceptual quality). Results demonstrate that the proposed techniques outperform existing methods, providing superior enhancement while minimizing artifacts and maintaining natural appearance. This research presents a robust framework for adaptive image enhancement, particularly effective in low-light conditions, making it valuable for applications requiring high precision and clarity.