Cargando…

Hyperspectral Image Classification with Capsule Network Using Limited Training Samples

Deep learning techniques have boosted the performance of hyperspectral image (HSI) classification. In particular, convolutional neural networks (CNNs) have shown superior performance to that of the conventional machine learning algorithms. Recently, a novel type of neural networks called capsule net...

Descripción completa

Detalles Bibliográficos
Autores principales: Deng, Fei, Pu, Shengliang, Chen, Xuehong, Shi, Yusheng, Yuan, Ting, Pu, Shengyan
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2018
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6165568/
https://www.ncbi.nlm.nih.gov/pubmed/30231574
http://dx.doi.org/10.3390/s18093153
_version_ 1783359867968290816
author Deng, Fei
Pu, Shengliang
Chen, Xuehong
Shi, Yusheng
Yuan, Ting
Pu, Shengyan
author_facet Deng, Fei
Pu, Shengliang
Chen, Xuehong
Shi, Yusheng
Yuan, Ting
Pu, Shengyan
author_sort Deng, Fei
collection PubMed
description Deep learning techniques have boosted the performance of hyperspectral image (HSI) classification. In particular, convolutional neural networks (CNNs) have shown superior performance to that of the conventional machine learning algorithms. Recently, a novel type of neural networks called capsule networks (CapsNets) was presented to improve the most advanced CNNs. In this paper, we present a modified two-layer CapsNet with limited training samples for HSI classification, which is inspired by the comparability and simplicity of the shallower deep learning models. The presented CapsNet is trained using two real HSI datasets, i.e., the PaviaU (PU) and SalinasA datasets, representing complex and simple datasets, respectively, and which are used to investigate the robustness or representation of every model or classifier. In addition, a comparable paradigm of network architecture design has been proposed for the comparison of CNN and CapsNet. Experiments demonstrate that CapsNet shows better accuracy and convergence behavior for the complex data than the state-of-the-art CNN. For CapsNet using the PU dataset, the Kappa coefficient, overall accuracy, and average accuracy are 0.9456, 95.90%, and 96.27%, respectively, compared to the corresponding values yielded by CNN of 0.9345, 95.11%, and 95.63%. Moreover, we observed that CapsNet has much higher confidence for the predicted probabilities. Subsequently, this finding was analyzed and discussed with probability maps and uncertainty analysis. In terms of the existing literature, CapsNet provides promising results and explicit merits in comparison with CNN and two baseline classifiers, i.e., random forests (RFs) and support vector machines (SVMs).
format Online
Article
Text
id pubmed-6165568
institution National Center for Biotechnology Information
language English
publishDate 2018
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-61655682018-10-10 Hyperspectral Image Classification with Capsule Network Using Limited Training Samples Deng, Fei Pu, Shengliang Chen, Xuehong Shi, Yusheng Yuan, Ting Pu, Shengyan Sensors (Basel) Article Deep learning techniques have boosted the performance of hyperspectral image (HSI) classification. In particular, convolutional neural networks (CNNs) have shown superior performance to that of the conventional machine learning algorithms. Recently, a novel type of neural networks called capsule networks (CapsNets) was presented to improve the most advanced CNNs. In this paper, we present a modified two-layer CapsNet with limited training samples for HSI classification, which is inspired by the comparability and simplicity of the shallower deep learning models. The presented CapsNet is trained using two real HSI datasets, i.e., the PaviaU (PU) and SalinasA datasets, representing complex and simple datasets, respectively, and which are used to investigate the robustness or representation of every model or classifier. In addition, a comparable paradigm of network architecture design has been proposed for the comparison of CNN and CapsNet. Experiments demonstrate that CapsNet shows better accuracy and convergence behavior for the complex data than the state-of-the-art CNN. For CapsNet using the PU dataset, the Kappa coefficient, overall accuracy, and average accuracy are 0.9456, 95.90%, and 96.27%, respectively, compared to the corresponding values yielded by CNN of 0.9345, 95.11%, and 95.63%. Moreover, we observed that CapsNet has much higher confidence for the predicted probabilities. Subsequently, this finding was analyzed and discussed with probability maps and uncertainty analysis. In terms of the existing literature, CapsNet provides promising results and explicit merits in comparison with CNN and two baseline classifiers, i.e., random forests (RFs) and support vector machines (SVMs). MDPI 2018-09-18 /pmc/articles/PMC6165568/ /pubmed/30231574 http://dx.doi.org/10.3390/s18093153 Text en © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Deng, Fei
Pu, Shengliang
Chen, Xuehong
Shi, Yusheng
Yuan, Ting
Pu, Shengyan
Hyperspectral Image Classification with Capsule Network Using Limited Training Samples
title Hyperspectral Image Classification with Capsule Network Using Limited Training Samples
title_full Hyperspectral Image Classification with Capsule Network Using Limited Training Samples
title_fullStr Hyperspectral Image Classification with Capsule Network Using Limited Training Samples
title_full_unstemmed Hyperspectral Image Classification with Capsule Network Using Limited Training Samples
title_short Hyperspectral Image Classification with Capsule Network Using Limited Training Samples
title_sort hyperspectral image classification with capsule network using limited training samples
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6165568/
https://www.ncbi.nlm.nih.gov/pubmed/30231574
http://dx.doi.org/10.3390/s18093153
work_keys_str_mv AT dengfei hyperspectralimageclassificationwithcapsulenetworkusinglimitedtrainingsamples
AT pushengliang hyperspectralimageclassificationwithcapsulenetworkusinglimitedtrainingsamples
AT chenxuehong hyperspectralimageclassificationwithcapsulenetworkusinglimitedtrainingsamples
AT shiyusheng hyperspectralimageclassificationwithcapsulenetworkusinglimitedtrainingsamples
AT yuanting hyperspectralimageclassificationwithcapsulenetworkusinglimitedtrainingsamples
AT pushengyan hyperspectralimageclassificationwithcapsulenetworkusinglimitedtrainingsamples