Recently, convolutional neural networks (CNNs) have achieved great improvements in single image dehazing and attained much attention in research. Most existing learning-based dehazing methods are not fully end-to-end, which still follow the traditional dehazing procedure: first estimate the medium transmission and the atmospheric light, then recover the haze-free image based on the atmospheric scattering model. However, in practice, due to lack of priors and constraints, it is hard to precisely estimate these intermediate parameters. Inaccurate estimation further degrades the performance of dehazing, resulting in artifacts, color distortion and insufficient haze removal. To address this, we propose a fully end-to-end Generative Adversarial Networks with Fusion-discriminator (FD-GAN) for image dehazing. With the proposed Fusion-discriminator which takes frequency information as additional priors, our model can generator more natural and realistic dehazed images with less color distortion and fewer artifacts. Moreover, we synthesize a large-scale training dataset including various indoor and outdoor hazy images to boost the performance and we reveal that for learning-based dehazing methods, the performance is strictly influenced by the training data. Experiments have shown that our method reaches state-of-the-art performance on both public synthetic datasets and real-world images with more visually pleasing dehazed results.

中文翻译：

---

FD-GAN：用于单图像去雾的具有融合鉴别器的生成对抗网络

最近，卷积神经网络（CNN）在单幅图像去雾方面取得了很大的进步，并在研究中受到了广泛关注。大多数现有的基于学习的去雾方法并不是完全端到端的，它们仍然遵循传统的去雾程序：首先估计介质透射和大气光，然后基于大气散射模型恢复无雾图像。然而，在实践中，由于缺乏先验和约束，很难精确估计这些中间参数。不准确的估计会进一步降低去雾的性能，导致伪影、颜色失真和去雾不足。为了解决这个问题，我们提出了一种具有融合鉴别器（FD-GAN）的完全端到端的生成对抗网络，用于图像去雾。使用将频率信息作为附加先验的所提出的融合鉴别器，我们的模型可以生成更自然、更逼真的去雾图像，颜色失真更少，伪影更少。此外，我们合成了一个包含各种室内和室外朦胧图像的大规模训练数据集以提高性能，并且我们发现对于基于学习的去雾方法，性能受到训练数据的严格影响。实验表明，我们的方法在公共合成数据集和真实世界图像上都达到了最先进的性能，并具有更令人愉悦的去雾结果。我们合成了一个包含各种室内和室外朦胧图像的大规模训练数据集以提高性能，我们发现对于基于学习的去雾方法，性能受到训练数据的严格影响。实验表明，我们的方法在公共合成数据集和真实世界图像上都达到了最先进的性能，并具有更令人愉悦的去雾结果。我们合成了一个包含各种室内和室外朦胧图像的大规模训练数据集以提高性能，并且我们发现对于基于学习的去雾方法，性能受到训练数据的严格影响。实验表明，我们的方法在公共合成数据集和真实世界图像上都达到了最先进的性能，并具有更令人愉悦的去雾结果。