This paper introduces a new approach for enhancing low-light images by utilizing coefficient bounds from a specific subclass of analytic functions, a concept rooted in geometric function theory. Our method is designed to adapt dynamically to different lighting conditions, ensuring effective image enhancement in both uniformly and non-uniformly illuminated environments. Specifically, we apply a convolution process for images that are evenly illuminated, while a power-law transformation is employed when illumination is non-uniform. This dual-method strategy allows us to overcome common challenges such as over-enhancement and under-enhancement, which are frequently encountered in traditional approaches. By adjusting to varying lighting conditions, our approach guarantees superior image quality across a range of scenarios. Through extensive experiments and rigorous comparisons using performance metrics, we demonstrate that our method consistently outperforms existing techniques, yielding significant improvements in both image clarity and contrast. These results confirm the effectiveness and adaptability of our approach in enhancing low-light images.
Read full abstract