Inland river image defogging based on optimized contrast enhancement
Bo Liu, Xiong Qing-guo · 2016 IEEE Information Technology, Networking, Electronic and Automation Control Conference · 2016
The fog images of Inland River often have low contrast. This paper introduces a optimized contrast enhancement algorithm. Firstly, divide the fog image by using the hierarchical searching method of quad-tree subdivision. Choose the brightest pixel as the atmospheric light in the final selected block. Then, combine the Mean Squared Error and the loss function. Next, get the optimal transmission value which can balance the contrast and the information loss. Then, fix the transmission with guide filter and adaptive window slide. Finally, restore the defogged image by solving the atmosphere attenuation model. Compared with the results obtained by several existing classical defogging methods, the experimental result shows that the proposed algorithm can effectively reduce the color distortion in sky region.