Hyperspectral Image Classification Based on Two-Channel Variational Autoencoder

doi:10.3778/j.issn.1002-8331.2008-0415

Abstract

Abstract: Aiming at the problems of low utilization efficiency of spatial and spectral features in hyperspectral image variational autoencoder（VAE） classification algorithm, a deep learning classification algorithm for hyperspectral images based on two-channel VAE is proposed. One-dimensional conditional variational autoencoder（CVAE） and two-dimensional channel-recurrent conditional variational autoencoders（CRCVAE） feature extraction frameworks are used to extract spectral and spatial features of hyperspectral images, respectively. These eigenvectors are superposed to form a spatial-spectral feature and classified by using the Softmax classifier. The analysis and verification are conducted on two hyperspectral data sets of the Indian pines and the Pavia University. The experimental results show that compared with other algorithms in this paper, the proposed algorithm improves at least 3.40, 2.75 and 3.57 percentage points in terms of overall classification accuracy, average classification accuracy and Kappa coefficient. The results show that the proposed algorithm has the highest classification accuracy and better visualization effect.

Key words: deep learning, hyperspectral image, classification, variational autoencoder, spatial-spectral feature

摘要： 针对现有高光谱图像变分自编码器（variational autoencoder，VAE）分类算法存在空间和光谱特征利用效率低的问题，提出一种基于双通道变分自编码器的高光谱图像深度学习分类算法。通过构建一维条件变分自编码器（conditional variational autoencoder，CVAE）特征提取框架和二维循环通道条件变分自编码（channel-recurrent conditional variational autoencoders，CRCVAE）特征提取框架分别提取高光谱图像的光谱特征和空间特征，将光谱特征向量和空间特征向量叠加形成空谱联合特征向量，将联合特征送入Softmax分类器中进行分类。在Indian pines和Pavia University两种高光谱数据集上进行了分析验证，实验结果显示，与其他算法相比，提出的算法在总分类精度、平均分类精度和Kappa系数等评价指标上至少提高了3.40、2.75和3.57个百分点，结果显示提出的算法得到了最高的分类精度和更好的可视化效果。

关键词: 深度学习, 高光谱图像, 分类, 变分自编码器, 空谱联合特征

LIU Zunxiong, SHI Yapeng, PENG Xiaoyu, WANG Yihong. Hyperspectral Image Classification Based on Two-Channel Variational Autoencoder[J]. Computer Engineering and Applications, 2022, 58(2): 244-251.

刘遵雄, 石亚鹏, 彭潇雨, 王毅宏. 基于双通道变分自编码器的高光谱图像分类[J]. 计算机工程与应用, 2022, 58(2): 244-251.

References

[1] 张号逵，李映，姜晔楠.深度学习在高光谱图像分类领域的研究现状与展望[J].自动化学报，2018，44（6）：961-977.
ZHANG H K，LI Y，JIANG Y N.Deep learning for hyperspectral imagery classification：the state of the art and prospects[J].Acta Automatica Sinica，2018，44（6）：961-977.
[2] LIU C Y，LI J，HE L，et al.Naive gabor networks for hyperspectral image classification[J].IEEE Transactions on Neural Networks and Learning Systems，2021，32（1）：376-390.
[3] JIA S，LIN Z J，DENG B，et al.Cascade superpixel regularized gabor feature fusion for hyperspectral image classification[J].IEEE Transactions on Neural Networks and Learning Systems，2020，31（5）：1638-1652.
[4] ZHAI H，ZHANG H，ZHANG L，et al.Cloud/shadow detection based on spectral indices for multi/hyperspectral optical remote sensing imagery[J].ISPRS Journal of Photogrammetry and Remote Sensing，2018，144（10）：235-253.
[5] WANG M W，WU C M，CHEN M L，et al.A band selection approach based on Lévy sine cosine algorithm and alternative distribution for hyperspectral image[J].Taylor & Francis，2020，41（9）：3429-3445.
[6] HSIEH T H，KIANG J F.Comparison of CNN algorithms on hyperspectral image classification in agricultural lands[J].Sensors，2020，20（6）：1734.
[7] NIKONOW W，RAMMLMAIR D，MEIMA J A，et al.Advanced mineral characterization and petrographic analysis by μ-EDXRF，LIBS，HSI and hyperspectral data merging[J].Mineralogy and Petrology，2019，113（3）：417-431.
[8] MIYOSHI G T，IMAI N N，TOMMASELLI A M G，et al.Radiometric block adjustment of hyperspectral image blocks in the Brazilian environment[J].International Journal of Remote Sensing，2018，39：15-16.
[9] SHIMONI M，HAELTERMAN R，PERNEEL C.Hypersectral imaging for military and security applications：combining myriad processing and sensing techniques[J].IEEE Geoscience and Remote Sensing，2019，7（2）：101-117.
[10] 崔颖，徐凯，陆忠军，等.主动学习策略融合算法在高光谱图像分类中的应用[J].通信学报，2018，39（4）：91-99.
CUI Y，XU K，LU Z J，et al.Combination strategy of active learning for hyperspectral images classification[J].Journal on Communications，2018，39（4）：91-99.
[11] 王立伟，李吉明，周国民，等.深度迁移学习在高光谱图像分类中的运用[J].计算机工程与应用，2019，55（5）：181-186.
WANG L W，LI J M，ZHOU G M，et al.Application of deep transfer learning in hyperspectral image classification[J].Computer Engineering and Applications，2019，55（5）：181-186.
[12] 毕晓君，周泽宇.基于双通道GAN的高光谱图像分类算法[J].光学学报，2019，39（10）：297-308.
BI X J，ZHOU Z Y.Hyperspectral image classification algorithm based on two-channel generative adversarial network[J].Acta Optica Sinica，2019，39（10）：297-308.
[13] CHEN Y，LIN Z，ZHAO X，et al.Deep learning based classification of hyperspectral data[J].IEEE J Sel Topics Appl Earth Observ Remote Sens，2014，7（6）：2094-2107.
[14] CHEN Y，ZHAO X，JIA X.Spectral-spatial classification of hyperspectral data based on deep belief network[J].IEEE J Sel Topics Appl Earth Observ Remote Sens，2015，8（6）：1-12.
[15] ZHU L，CHEN Y S，GHAMISI P，et al.Generative adversarial networks for hyperspectral image classification[J].IEEE Transactions on Geoscience and Remote Sensing，2018，56（9）：5046-5063.
[16] HANG R，LIU Q，HONG D，et al.Cascaded recurrent neural networks for hyperspectral image classification[J].IEEE Transactions on Geoscience and Remote Sensing，2019，57（8）：5384-5394.
[17] 佘海龙，解山娟，邹静洁.标准分数降维的3D-CNN 高光谱遥感图像分类[J].计算机工程与应用，2021，57（4）：169-175.
SHE H L，XIE S J，ZOU J J.3D-CNN with standard score based dimensionality reduction for hyperspectral remote sensing images classification[J].Computer Engineering and Applications，2021，57（4）：169-175.
[18] TUIA D，VOLPI M，TROLLIET M，et al.Semisupervised manifold alignment of multimodal remote sensing images[J].IEEE Transactions on Geoscience and Remote Sensing，2014，52（12）：7708-7720.
[19] SAMAT A，LI J，LIN C，et al.Edge gradient-based active learning for hyperspectral image classification[J].IEEE Geoscience and Remote Sensing Letters，2020，17（9）：1588-1592.
[20] ZHONG Y，ZHANG L，GONG W.Unsupervised remote sensing image classification using an artificial immune network[J].Remote Sensing，2011，32（19）：5461-5483.
[21] HUO H Y，GUO J F，LI Z L.Hyperspectral image classification for land cover based on an improved interval Type-II fuzzy C-means approach[J].Sensors，2018，18（2）：363.
[22] LI A L，QIN A Y，SHANG Z W，et al.Spectral-spatial sparse subspace clustering based on three-dimensional edge-preserving filtering for hyperspectral image[J].International Journal of Pattern Recognition and Artificial Intelligence，2019，33（3）：1955003.
[23] BODIN E，MALIK I，EK C H，et al.Nonparametric inference for autoencoding variational Bayes[J].arXiv：1712.
06536v1，2017.
[24] KINGMA D P，WELLING M.Auto-encoding variational Bayes[C]//Proceedings of the International Conference on Learning Representations，2014.
[25] 于纯妍，赵猛，宋梅萍，等.基于目标约束与谱空迭代的高光谱图像分类方法[J].光学学报，2018，38（6）：319-329.
YU C Y，ZHAO M，SONG M P，et al.Hyperspectral image classification method based on targets constraint and spectral-spatial iteration[J].Acta Optica Sinica，2018，38（6）：319-329.