Proxy-Based Deep Learning Framework for Spectral–Spatial Hyperspectral Image Classification: Efficient and Robust
- 8 February 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 60, 1-15
- https://doi.org/10.1109/tgrs.2021.3054008
Abstract
Deep convolutional networks have been extensively deployed in hyperspectral image (HSI) classification. Reaching for high accuracy, the existing deep-learning-based methods commonly deepen or widen their networks for better performance, which brings higher computational complexity and the risk of overfitting. Although the introduction of the residual module and batch-normalization reduces the generalization degradation in complex networks, the mainstream methods still suffer from low robustness to the noise. To tackle these issues, a compact proxy-based deep learning framework is proposed to perform highly accurate HSI classification with superb efficiency and robustness. In this article: 1) novel deep proxies are integrated to replace the dense classifier layers in conventional networks, which represents specific classes in deep embedding space and enables fast and reliable convergence; 2) the proxy-based feature embedding is studied in distance metric and similarity metric, and compatible dual-metric loss functions are designed for further optimized embedding distribution, which leads to more robust generalization; and 3) state-of-the-art performance and robustness are demonstrated by the proposed framework on mainstream HSI data sets with the minimal network scale and time complexity.Keywords
Funding Information
- National Key Research and Development Project (2020YFB2103902)
- National Science Fund for Distinguished Young Scholars (61825603)
- Key Program of National Natural Science Foundation of China (61632018)
- National Natural Science Foundation of China (62001397)
- Natural Science Basic Research Program of Shaanxi (2020JQ-212)
- Open-Ended Foundation of the National Radar Signal Processing Laboratory (61424010207)
This publication has 55 references indexed in Scilit:
- Structured Priors for Sparse-Representation-Based Hyperspectral Image ClassificationIEEE Geoscience and Remote Sensing Letters, 2013
- Stacked Convolutional Auto-Encoders for Hierarchical Feature ExtractionLecture Notes in Computer Science, 2011
- An introduction to hyperspectral imaging and its application for security, surveillance and target acquisitionThe Imaging Science Journal, 2010
- Random Forests for land cover classificationPattern Recognition Letters, 2006
- Hyperspectral vegetation indices and novel algorithms for predicting green LAI of crop canopies: Modeling and validation in the context of precision agricultureRemote Sensing of Environment, 2004
- Comparison of airborne hyperspectral data and eo-1 hyperion for mineral mappingIEEE Transactions on Geoscience and Remote Sensing, 2003
- Evaluation of prototype learning algorithms for nearest-neighbor classifier in application to handwritten character recognitionPattern Recognition, 2001
- Hyperspectral Vegetation Indices and Their Relationships with Agricultural Crop CharacteristicsRemote Sensing of Environment, 2000
- Nearest prototype classification: clustering, genetic algorithms, or random search?IEEE Transactions on Systems, Man and Cybernetics, Part C (Applications and Reviews), 1998
- Imaging Spectrometry for Earth Remote SensingScience, 1985