Dual Multi-scale Dehazing Network

Single-image haze removal is a challenging ill-posed problem. Recently, methods based on training synthetic data have achieved good dehazing results. However, we note that these can be further improved. A novel deep learning-based method proposed to obtain better-dehazed result for single-image in this paper. Specially, propose dual multi-scale network learn the knowledge from synthetical data. The coarse designed capture large variety of objects, and then fine blocks are small objects at each scale. To show effectiveness method, perform experiments dataset real hazy images. Extensive experimental results outperforms state-of-the-art methods.