Capsule Networks for Hyperspectral Image Classification
Top Cited Papers
- 25 October 2018
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 57 (4) , 2145-2160
Abstract
Convolutional neural networks (CNNs) have recently exhibited an excellent performance in hyperspectral image classification tasks. However, the straightforward CNN-based network architecture still finds obstacles when effectively exploiting the relationships between hyperspectral imaging (HSI) features in the spectral-spatial domain, which is a key factor to deal with the high level of complexity present in remotely sensed HSI data. Despite the fact that deeper architectures try to mitigate these limitations, they also find challenges with the convergence of the network parameters, which eventually limit the classification performance under highly demanding scenarios. In this paper, we propose a new CNN architecture based on spectral-spatial capsule networks in order to achieve a highly accurate classification of HSIs while significantly reducing the network design complexity. Specifically, based on Hinton's capsule networks, we develop a CNN model extension that redefines the concept of capsule units to become spectral-spatial units specialized in classifying remotely sensed HSI data. The proposed model is composed by several building blocks, called spectral-spatial capsules, which are able to learn HSI spectral-spatial features considering their corresponding spatial positions in the scene, their associated spectral signatures, and also their possible transformations. Our experiments, conducted using five well-known HSI data sets and several state-of-the-art classification methods, reveal that our HSI classification approach based on spectral-spatial capsules is able to provide competitive advantages in terms of both classification accuracy and computational time.Keywords
Funding Information
- Consejería de Educación y Empleo, Junta de Extremadura
- Consejería de Educación y Empleo, Junta de Extremadura (GR15005)
- Generalitat Valenciana (APOSTD/2017/007)
- Spanish Ministry of Economy (ESP2016-79503-C2-2-P, TIN2015-63646-C5-5-R)
- National Natural Science Foundation of China (61771496)
- National Key Research and Development Program of China (2017YFB0502900)
- Natural Science Foundation of Guangdong Province (2016A030313254)
This publication has 44 references indexed in Scilit:
- Deep learningNature, 2015
- Deep Learning-Based Classification of Hyperspectral DataIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014
- Vehicle Detection in Satellite Images by Hybrid Deep Convolutional Neural NetworksIEEE Geoscience and Remote Sensing Letters, 2014
- Mixed Pooling for Convolutional Neural Networks2014
- Advances in Hyperspectral Image Classification: Earth Monitoring with Statistical Learning MethodsIEEE Signal Processing Magazine, 2013
- Transforming Auto-Encoders2011
- A review on hyperspectral remote sensing for homogeneous and heterogeneous forest biodiversity assessmentInternational Journal of Remote Sensing, 2010
- Hyperspectral image data analysisIEEE Signal Processing Magazine, 2002
- Wavelets, vision and the statistics of natural scenesPhilosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 1999
- The pixel: A snare and a delusionInternational Journal of Remote Sensing, 1997