Single image dehazing based on maximizing local contrast
Huiying Dong, Changkai Zhao · 2015
In order to restore visibility from a single degraded image, we propose a fast and efficient method based on local contrast maximization and atmospheric scattering model. According to the characteristics of outdoor imaging and the theory of graph cuts, the original method based on local contrast maximization was improved and optimized. We processed a number of degraded outdoor images to verify the effectiveness of our algorithm. The experimental results show that this method compared with the original has an advantage of processing speed, and it can be able to get a better dehazing output. Our method remarkably improves the stability and reliability of outdoor imaging devices.