Speaker
Description
This article presents a comprehensive study of classical and modern approaches to enhancing images captured in low-light conditions. An adaptive ensemble method is proposed that combines four classical image processing algorithms and automatically weights them based on local quality. The developed system uses CLAHE, gamma correction, the Retinex algorithm, and automatic contrast stretching as its basic components. A comparative analysis of the developed method with the modern Deep Learning approach, Zero-DCE was conducted on the real-world LOL dataset. Additionally, the effects of four types of noise on the effectiveness of image enhancement were experimentally investigated. The experimental results show that the proposed classical ensemble approach achieves a PSNR of 17.91 dB, outperforming the machine learning model by 14.2 percent, while the method does not require a training phase on datasets or GPU computations.
Keywords: image enhancement, low-light enhancement, ensemble methods, CLAHE, Zero-DCE, adaptive weighting, image processing.