Image Dehaze Method Using Depth Map Estimation Network Based on Atmospheric Scattering Model

摘要

The purpose of dehaze is to remove haze effects for image enhancement. In previous typical researches for dehaze using deep learning, the transmission map is estimated directly through the training process. In this paper, we propose to obtain the transmission map indirectly from the depth map, which is estimated using deep learning framework. The haze removal is performed from the transmission map based on the atmospheric scattering model as the existing method. The experimental results show that the proposed method can give improved dehaze results for some images with relatively dark conditions.