A new single image dehazing method with MSRCR algorithm
Jinbao Wang, Ning He, Ke Lu · 2015
In this paper, we propose a new single image dehazing method using a Multi-Scale Retinex with Color Restoration algorithm. The overall dehazing process involves three parts, the atmospheric light value calculation, transmission map estimation, and optimization after dehazing. In addition to a subjective evaluation, image quality was also evaluated objectively using two indicators, Average Gradient, and Information Entropy, in order to consider the detail of the image texture, and image dynamic range. The experimental results show that our algorithm can effectively improve the image quality degraded by foggy weather, is fast, and can retain image details effectively.