A novel encoder-decoder network with guided transmission map for single image dehazing

A novel Encoder-Decoder Network with Guided Transmission Map (EDN-GTM) for single image dehazing scheme is proposed in this paper. The EDN-GTM takes conventional RGB hazy conjunction its transmission map estimated by adopting dark channel prior as the inputs of network. utilizes U-Net segmentation core network and various modifications including spatial pyramid pooling module Swish activation to achieve state-of-the-art performance. Experiments on benchmark datasets show that outperforms most traditional deep learning-based schemes terms PSNR SSIM metrics. furthermore proves applicability object detection problems. Specifically, when applied an preprocessing tool driving detection, can efficiently remove haze significantly improve accuracy 4.73% mAP measure.