A Review of Modern Low Light Image Enhancement Techniques

Enhancing low-light images is a major challenge in computer vision. It suffersfrom modest contrast, unclear details, and intensive noise. Low-w light image enhancement aimsed to enhance the quality of the image that is captured under imperfect lighting and restore the image with a high-accuracy colour distributioned with standard lighting. This paper discusses the evolution of techniques used to improve image quality, starting from traditional methods such as dehazingg, histogram equalisation (HE), gamma correction (GC), and Retinex theory methods to modern techniques that rely on deep learning networks based on specially trained models for low- light image enhancement.t Dynamic and restoring lost details, for example, CNNs, GANs, and U-Net.

Low light image enhancement (LLIE), Datasets, traditional techniques, deep learning methods.

10.52113/2/12.01.2025/30-48