ABSTRACT

In this letter, we present a deep-learning-based methodology for recovering hyperspectral images (HSIs) distorted by Gaussian and impulsive noise. This work makes the following contribution: To begin with, the Wasserstein Generative Adversarial Network (WGAN) is used to mitigate the effects of vanishing gradient and mode collapse that can occur when training a vanilla GAN. Secondly, data are passed via three distinct pathways in a parallel ensemble to promote multiscale feature extraction. Normal and multiscale dilated 3D convolutions are utilized to train the model in each pair of parallel paths. Thirdly, features are recovered following data permutation across three different spatial planes (viz. $\mathrm{x}\mathrm{y},\mathrm{y}\mathrm{z}$, and $\mathrm{x}\mathrm{z}$ planes) and after passing through parallel convolutional blocks; to promote spatio-spectral similarity within and across the different layers of the HSI data. Fourthly, by adopting Structural Similarity (SSIM) as the content loss, the issue of loss in resolution encountered during adversarial training is mitigated. Finally, the incorporation of 3D depth-wise separable convolution and batch re-normalization (BRN) solves the major issue of computational burden encountered while processing HSI data. Extensive experimental evaluation on synthetically corrupted data and real HSI data (obtained from real hyperspectral sensors) under various degradation conditions suggests that the aforementioned denoising approach could be used in real time.

摘要

在这封信中，我们提出了一种基于深度学习的方法，用于恢复被高斯和脉冲噪声扭曲的高光谱图像 (HSI)。这项工作做出了以下贡献：首先，Wasserstein 生成对抗网络 (WGAN) 用于减轻训练 vanilla GAN 时可能发生的梯度消失和模式崩溃的影响。其次，数据通过并行集成中的三个不同路径传递，以促进多尺度特征提取。正常和多尺度扩张 3D 卷积用于在每对平行路径中训练模型。第三，在三个不同空间平面（即。$\mathrm{X}\mathrm{是的},\mathrm{是的}\mathrm{z}$， 和$\mathrm{X}\mathrm{z}$平面）和通过并行卷积块之后；促进 HSI 数据不同层内和跨层的空间光谱相似性。第四，通过采用结构相似性（SSIM）作为内容损失，缓解了对抗训练期间遇到的分辨率损失问题。最后，3D 深度可分离卷积和批量重新归一化 (BRN) 的结合解决了处理 HSI 数据时遇到的计算负担的主要问题。在各种退化条件下对综合损坏数据和真实 HSI 数据（从真实高光谱传感器获得）的广泛实验评估表明，上述去噪方法可以实时使用。