Histogram Matching and Fusion for Effective Low-Light Image Enhancement
Shaffa K. Kokro, Elijah Githinji Mwangi, George Kamucha · 2024
The enhancement of low-light images is critical for many applications, including surveillance, medical imaging, and astronomy. Existing approaches, however, frequently exhibit colour distortions, over-saturation, or other abnormalities that drastically reduce the low-light image fidelity. In this paper, we propose a novel image enhancement algorithm that addresses these challenges by leveraging techniques such as histogram matching and the discrete wavelet transform image fusion. We then employ the classical unsharp masking algorithm to further improve the details and edges in the resulting image. Computer simulations on benchmark datasets show the efficacy of our method in both objective quality and subjective visual inspection. Compared to existing approaches, our method achieved high PSNR and SSIM quality, demonstrating the potential of our algorithm for low-light image enhancement applications.