<p>The uneven illumination effect is associated with brightness variations, where some image areas appear darker or brighter than others. It occurs due to sensor or lens limitations, nonuniform lighting, shadowing, and reflection. Despite technological development, this effect remains visible in captured images, making them appear deficient in illumination. Image processing is helpful in various applications and is utilized to handle this effect. When the classical single-scale Retinex (SSR) was tested with many different images, it introduced incorrect illumination, color misrepresentation, halos, and deficient contrast. Adequate correction of uneven illumination is challenging, as it relates to important aspects such as preserving the bright areas from being massively amplified, brightening the dark areas without introducing flaws, and providing adequate contrast and color representation. Various algorithms have been introduced to solve the illumination issue, but many have been provisional. Hence, a refined SSR (RSSR) algorithm is developed in this study to correct the uneven illumination effect and avoid the shortcomings of the classical SSR in the resulting images. The RSSR employs a new model to compute the reflectance and applies three more models to refine its details, processing the V channel only in the HSV color space. The RSSR is experimented with real-world, unevenly illuminated images along with performance assessments by four specialized measures. Also, comparisons are made with ten dissimilar algorithms. The results revealed the RSSR's success in correcting the illumination of various images and outperforming many existing algorithms in terms of rapidity, illumination quality, contrast, and structural naturalness.</p>