An Efficient Algorithm Based On The Fast Fuzzy Theory For Image And Video Dehazing
Rong Zheng,Haiyang Zhou,Lei Zhang,Lin Yan,Feihong Yu
DOI: https://doi.org/10.1117/12.2242617
2016-01-01
Abstract:Influenced by the climatic conditions, such as haze, there exist problems of weak visibility and low contrast for the images and videos captured outdoors. Recently, an effective image haze removal method based on dark channel prior has been proposed. However, the brightness of the result is usually not as bright as the atmospheric light, that makes the whole image looks dim. Besides, the execution speed of this method is slow. As a result, it cannot be applied to the situations with high real-time requirements, such as video streams. In order to solve these problems, an efficient algorithm for image and video dehazing is proposed in this paper. Firstly, the transmission map of hazy image based on the fast fuzzy theory is calculated. Then, according to the statistical principle of dark channel prior and the atmospheric scattering model, the haze-free image under ideal illumination can be restored successfully. Large number of experimental results have shown that, the proposed algorithm can obtain better haze-free results for single image compared with the previous method. More importantly, the execution efficiency has been improved greatly. As a result, video steaming can also be dehazed in real-time so as to meet the occasions with much requirements of industry.