Spatial Revising Variational Autoencoder-Based Feature Extraction Method for Hyperspectral Images
Spatial Revising Variational Autoencoder-Based Feature Extraction Method for Hyperspectral Images
复制标题
基于空间修正变分自编码器的高光谱图像特征提取方法
DOI：
10.1109/tgrs.2020.2997835
复制
发表时间：
2020-06
影响因子：
8.2
通讯作者：
Shen Yi
中科院分区：
文献类型：
--
作者：
Yu Wenbo;Zhang Miao;Shen Yi
高维高光谱图像往往会增加计算量，给图像处理带来挑战。深度学习模型在各种图像处理领域取得了非凡的成功，这些模型可以有效地提高分类性能。在充分提取丰富的光谱信息，如空间和光谱信息的组合仍然存在相当大的挑战。提出了一种基于空间修正变分自编码器（$U_{\text {Hfe}}\text {SRVAE}$）的无监督高光谱特征提取方法。该方法的核心思想是通过设计的网络从多个方面提取空间特征，对获得的光谱特征进行修正。多层编码器提取光谱特征，然后，潜在的空间向量生成从获得的均值和标准差。分别采用多层卷积神经网络和长短期记忆网络提取基于局部感知和序列感知的空间特征，并对得到的均值向量进行修正。此外，所提出的损失函数保证了从同一相邻区域获得的各种潜在空间特征的概率分布的一致性。在三个公开的高光谱数据集上进行了实验，实验结果表明，$U_{\text {Hfe}}\text {SRVAE}$取得了较好的分类效果。针对高光谱图像的特点，设计了空间特征提取模型和深层声发射模型的组合，提高了该方法的性能。
Hyperspectral image with high dimensionality always increases the computational consumption, which challenges image processing. Deep learning models have achieved extraordinary success in various image processing domains, which are effective to improve classification performance. There remain considerable challenges in fully extracting abundant spectral information, such as the combination of spatial and spectral information. In this article, a novel unsupervised hyperspectral feature extraction architecture based on spatial revising variational autoencoder (AE) ( $U_{\text {Hfe}}\text {SRVAE}$ ) is proposed. The core concept of this method is extracting spatial features via designed networks from multiple aspects for the revision of the obtained spectral features. Multilayer encoder extracts spectral features, and then, latent space vectors are generated from the obtained means and standard deviations. Spatial features based on local sensing and sequential sensing are extracted using multilayer convolutional neural networks and long short-term memory networks, respectively, which can revise the obtained mean vectors. Besides, the proposed loss function guarantees the consistency of the probability distributions of various latent spatial features, which obtained from the same neighbor region. Several experiments are conducted on three publicly available hyperspectral data sets, and the experimental results show that $U_{\text {Hfe}}\text {SRVAE}$ achieves better classification results compared with comparison methods. The combination of spatial feature extraction models and deep AE models is designed based on the unique characteristics of hyperspectral images, which contributes to the performance of this method.
登录
查看更多内容
DOI：
10.1007/978-981-10-7299-4_48
发表时间：
2017-10
期刊：
--
影响因子：
--
作者：
F. Zhou;Renlong Hang;Qingshan Liu;Xiaotong Yuan
通讯作者：
F. Zhou;Renlong Hang;Qingshan Liu;Xiaotong Yuan
DOI：
--
发表时间：
2001
期刊：
--
影响因子：
--
作者：
P. Belhumeur;D. Kriegman
通讯作者：
P. Belhumeur;D. Kriegman
DOI：
10.1007/bfb0015522
发表时间：
1996-04
期刊：
--
影响因子：
--
作者：
P. Belhumeur;J. Hespanha;D. Kriegman
通讯作者：
P. Belhumeur;J. Hespanha;D. Kriegman
DOI：
10.1016/j.procs.2018.03.048
发表时间：
2017
期刊：
--
影响因子：
--
作者：
Yanhui Guo-;Siming Han;Han Cao;Yu Zhang;Qian Wang
通讯作者：
Yanhui Guo-;Siming Han;Han Cao;Yu Zhang;Qian Wang
DOI：
--
发表时间：
2002-12
期刊：
ArXiv
影响因子：
--
作者：
Zhenyue Zhang;Hongyuan Zha
通讯作者：
Zhenyue Zhang;Hongyuan Zha