The Improved Retinex Algorithm Based on the HSV Color Space: Achieving Efficient Low-Light Image Enhancement and Color Fidelity

To address the prevalent issues of insufficient brightness, uneven illumination, low contrast, and blurred details in low-light environment images, we have devised a groundbreaking image enhancement technique. This technique ingeniously integrates the HSV color model with the Retinex theory. Initially, we convert the low-light image from the RGB color space to the HSV domain, enabling more precise adjustments to image properties. Subsequently, within the HSV framework, we utilize an advanced iteration of the Retinex algorithm, augmented with adaptive gamma correction and CLAHE (Contrast Limited Adaptive Histogram Equalization) techniques, specifically designed to optimize the luminance channel of the image. This enhancement notably boosts the brightness of low-light images.Further, by employing bilateral filtering and linear stretching techniques, we enhance the saturation component of the image, ensuring vivid colors. As the final stage, we transform the enhanced image back to the RGB color format and apply automated white balancing mechanisms to accurately restore the true hues of objects within the image. Experimental evaluations conclusively show that our algorithm effectively elevates the quality of low-light images, yielding superior visual outcomes.