Fine-Grained Classification of Hyperspectral Imagery Based on Deep Learning
Hyperspectral remote sensing obtains abundant spectral and spatial information of the observed object simultaneously. It is an opportunity to classify hyperspectral imagery (HSI) with a fine-grained manner. In this study, the fine-grained classification of HSI, which contains a large number of class...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-11-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/11/22/2690 |
| _version_ | 1819002317980041216 |
|---|---|
| author | Yushi Chen Lingbo Huang Lin Zhu Naoto Yokoya Xiuping Jia |
| author_facet | Yushi Chen Lingbo Huang Lin Zhu Naoto Yokoya Xiuping Jia |
| author_sort | Yushi Chen |
| collection | DOAJ |
| description | Hyperspectral remote sensing obtains abundant spectral and spatial information of the observed object simultaneously. It is an opportunity to classify hyperspectral imagery (HSI) with a fine-grained manner. In this study, the fine-grained classification of HSI, which contains a large number of classes, is investigated. On one hand, traditional classification methods cannot handle fine-grained classification of HSI well; on the other hand, deep learning methods have shown their powerfulness in fine-grained classification. So, in this paper, deep learning is explored for HSI supervised and semi-supervised fine-grained classification. For supervised HSI fine-grained classification, densely connected convolutional neural network (DenseNet) is explored for accurate classification. Moreover, DenseNet is combined with pre-processing technique (i.e., principal component analysis or auto-encoder) or post-processing technique (i.e., conditional random field) to further improve classification performance. For semi-supervised HSI fine-grained classification, a generative adversarial network (GAN), which includes a discriminative CNN and a generative CNN, is carefully designed. The GAN fully uses the labeled and unlabeled samples to improve classification accuracy. The proposed methods were tested on the Indian Pines data set, which contains 33,3951 samples with 52 classes. The experimental results show that the deep learning-based methods provide great improvements compared with other traditional methods, which demonstrate that deep models have huge potential for HSI fine-grained classification. |
| first_indexed | 2024-12-20T23:03:12Z |
| format | Article |
| id | doaj.art-4716d2a725be485f9751a38725da9d6a |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-12-20T23:03:12Z |
| publishDate | 2019-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-4716d2a725be485f9751a38725da9d6a2022-12-21T19:23:56ZengMDPI AGRemote Sensing2072-42922019-11-011122269010.3390/rs11222690rs11222690Fine-Grained Classification of Hyperspectral Imagery Based on Deep LearningYushi Chen0Lingbo Huang1Lin Zhu2Naoto Yokoya3Xiuping Jia4School of Electronics and Information Engineering, Harbin Institute of Technology, Harbin 150001, ChinaSchool of Electronics and Information Engineering, Harbin Institute of Technology, Harbin 150001, ChinaSchool of Electronics and Information Engineering, Harbin Institute of Technology, Harbin 150001, ChinaRIKEN Center for Advanced Intelligence Project, Tokyo 103-0027, JapanSchool of Engineering and Information Technology, The University of New South Wales, Canberra, ACT 2600, AustraliaHyperspectral remote sensing obtains abundant spectral and spatial information of the observed object simultaneously. It is an opportunity to classify hyperspectral imagery (HSI) with a fine-grained manner. In this study, the fine-grained classification of HSI, which contains a large number of classes, is investigated. On one hand, traditional classification methods cannot handle fine-grained classification of HSI well; on the other hand, deep learning methods have shown their powerfulness in fine-grained classification. So, in this paper, deep learning is explored for HSI supervised and semi-supervised fine-grained classification. For supervised HSI fine-grained classification, densely connected convolutional neural network (DenseNet) is explored for accurate classification. Moreover, DenseNet is combined with pre-processing technique (i.e., principal component analysis or auto-encoder) or post-processing technique (i.e., conditional random field) to further improve classification performance. For semi-supervised HSI fine-grained classification, a generative adversarial network (GAN), which includes a discriminative CNN and a generative CNN, is carefully designed. The GAN fully uses the labeled and unlabeled samples to improve classification accuracy. The proposed methods were tested on the Indian Pines data set, which contains 33,3951 samples with 52 classes. The experimental results show that the deep learning-based methods provide great improvements compared with other traditional methods, which demonstrate that deep models have huge potential for HSI fine-grained classification.https://www.mdpi.com/2072-4292/11/22/2690convolutional neural network (cnn)deep learninggenerative adversarial network (gan)hyperspectral imagery classificationsemi-supervised classification |
| spellingShingle | Yushi Chen Lingbo Huang Lin Zhu Naoto Yokoya Xiuping Jia Fine-Grained Classification of Hyperspectral Imagery Based on Deep Learning Remote Sensing convolutional neural network (cnn) deep learning generative adversarial network (gan) hyperspectral imagery classification semi-supervised classification |
| title | Fine-Grained Classification of Hyperspectral Imagery Based on Deep Learning |
| title_full | Fine-Grained Classification of Hyperspectral Imagery Based on Deep Learning |
| title_fullStr | Fine-Grained Classification of Hyperspectral Imagery Based on Deep Learning |
| title_full_unstemmed | Fine-Grained Classification of Hyperspectral Imagery Based on Deep Learning |
| title_short | Fine-Grained Classification of Hyperspectral Imagery Based on Deep Learning |
| title_sort | fine grained classification of hyperspectral imagery based on deep learning |
| topic | convolutional neural network (cnn) deep learning generative adversarial network (gan) hyperspectral imagery classification semi-supervised classification |
| url | https://www.mdpi.com/2072-4292/11/22/2690 |
| work_keys_str_mv | AT yushichen finegrainedclassificationofhyperspectralimagerybasedondeeplearning AT lingbohuang finegrainedclassificationofhyperspectralimagerybasedondeeplearning AT linzhu finegrainedclassificationofhyperspectralimagerybasedondeeplearning AT naotoyokoya finegrainedclassificationofhyperspectralimagerybasedondeeplearning AT xiupingjia finegrainedclassificationofhyperspectralimagerybasedondeeplearning |