Image dehazing using dark channels with global transmission

This paper presents a new single image dehazing algorithm using dark channels with global transmission. The existing algorithms using dark channels cannot deal well with the scene in which objects are inherently similar to the atmospheric light and no shadow is cast on them. In order to eliminate the limitation, we use the positive depth of field prior and the global dynamic template instead of local patch to estimate the transmission map. To solve the problem caused by the hanging objects, the hanging region is calculated to amend the transmission map. The experimental results show that the proposed algorithm performs well in the image with large white area and hanging region.