Fusion of Heterogeneous Adversarial Networks for Single Image Dehazing
Fusion of Heterogeneous Adversarial Networks for Single Image Dehazing
复制标题
DOI：
10.1109/tip.2020.2975986
复制
发表时间：
2020-01-01
影响因子：
10.6
通讯作者：
Ko, Hanseok
中科院分区：
文献类型：
--
作者：
Park, Jaihyun;Han, David K.;Ko, Hanseok
在本文中，我们提出了一种新的图像去雾方法。典型的去雾深度学习模型是在成对的合成室内数据集上训练的。因此，这些模型对于室内图像去雾可能是有效的，但对于室外图像则不那么有效。我们提出了一种基于异构生成对抗网络（GAN）的方法，该方法由用于生成模糊清晰图像的周期一致生成对抗网络（CycleGAN）和用于保留纹理细节的条件生成对抗网络（cGAN）组成。我们在融合网络的训练中引入了一种新的损失函数，以最大限度地减少GAN生成的伪影，恢复细节，并保留颜色分量。这些网络通过卷积神经网络（CNN）融合以生成去雾图像。大量的实验表明，该方法显着优于国家的最先进的方法在合成和现实世界的朦胧图像。
In this paper, we propose a novel image dehazing method. Typical deep learning models for dehazing are trained on paired synthetic indoor dataset. Therefore, these models may be effective for indoor image dehazing but less so for outdoor images. We propose a heterogeneous Generative Adversarial Networks (GAN) based method composed of a cycle-consistent Generative Adversarial Networks (CycleGAN) for producing haze-clear images and a conditional Generative Adversarial Networks (cGAN) for preserving textural details. We introduce a novel loss function in the training of the fused network to minimize GAN generated artifacts, to recover fine details, and to preserve color components. These networks are fused via a convolutional neural network (CNN) to generate dehazed image. Extensive experiments demonstrate that the proposed method significantly outperforms the state-of-the-art methods on both synthetic and real-world hazy images.