Dark channel prior algorithm for bright regions
Cui Bao-xi · Shenyang Gongye Daxue xuebao · 2015
In order to solve the problem that the restored images will exhibit serious color distortion when the haze images in the bright regions with large area are processed with dark channel prior algorithm,a transmission correction scheme based on pixels was proposed. Based on the analysis for both principle of dark channel prior haze removal and color distortion cause of restored images,the derivation was performed for the haze removal model. Through using the numerical proximity between dark channel and atmospheric light as well as the numerical proximity between channels as the decision mechanism of bright regions,a more universal method of calculating the transmission was obtained. For the bright regions,the error estimation of transmission was corrected with the tolerance mechanism,while the original transmission estimation method was still used in the non-bright regions. The improved algorithm was firstly applied to the clearness of snowimages. The experimental results showthat the haze removal images are clear and nature with the improved algorithm,and the color distortion can be effectively corrected.