Enlightening the Dark: How AI Becomes Humanity's Digital Night Vision Device

To address image enhancement under insufficient lighting conditions, this paper proposes an algorithm based on the Retinex theory for low-light image enhancement by converting the RGB color space to HSV. The primary objective of enhancing low-light images is to improve brightness, which is achieved through the following steps: (1) converting images from RGB to HSV color space; (2) enhancing the V-channel (value component) in HSV space to boost brightness while suppressing noise; and (3) reconstructing images back to RGB space. Experimental results demonstrate that the proposed method effectively reduces noise, enhances visual quality, and improves brightness compared to raw images. However, the runtime remains a challenge for practical applications. Subsequently, we investigate contrast, saturation, and contour enhancements in the processed images. To further optimize performance, we propose an AI-driven analytical approach that integrates machine learning techniques to reduce processing time while maintaining or improving imaging quality.