Neural Computing and Applications

, Volume 31, Issue 12, pp 8997–9012 | Cite as

Convolutional neural network for spectral–spatial classification of hyperspectral images

  • Hongmin Gao
  • Yao Yang
  • Chenming LiEmail author
  • Xiaoke Zhang
  • Jia Zhao
  • Dan Yao
Original Article


Hyperspectral images (HSIs) have great potential in military reconnaissance, land use, marine monitoring and many other fields. In recent years, the convolutional neural network (CNN) has been successfully used to classify hyperspectral data and achieved remarkable performance. However, the limited labeled samples of HSI lead to the small sample size problem, which remains the major challenge for CNN-based HSI classification. Besides, most CNN models have large number of parameters needed to be learned, which cause high computational cost. To address the aforementioned two issues, a novel CNN-based HSI classification method is proposed. The proposed classification method has several distinguishing characteristics. First, the proposed method can robustly extract spectral and spatial features of the HSI simultaneously. Second, in the proposed CNN architecture, all convolution layers are 1 × 1 convolution layer except the first one, which can greatly reduce the number of network parameters, thus accelerating the training and testing process. Third, a small convolution and feature reuse (SC-FR) module is developed. The SC-FR module is composed of two composite layers and each composite layer consists of two cascaded 1 × 1 convolution layers. Through cross-layer connecting, the input and output features of each composite layer are concatenated and passed to the next convolution layer, thus achieving feature reuse mechanism. Cross-layer connection increases information flow and the utilization rate of middle-level features, which enhances the generalization performance of CNN effectively. Experimental results on three benchmark HSIs demonstrate the competitive superiority of the proposed method over several state-of-the-art HSI classification methods, especially when training samples are limited.


Convolutional neural network Hyperspectral image classification Small sample size problem Small convolution 



This work was supported by National Natural Science Foundation of China (No. 61701166), National Key R&D Program of China (No. 2018YFC1508106), China Postdoctoral Science Foundation (No. 2018M632215), Fundamental Research Funds for the Central Universities (No. 2018B16314), Science Fund for Distinguished Young Scholars of Jiangxi Province under Grant (No. 2018ACB21029), Young Elite Scientists Sponsorship Program by CAST (No. 2017QNRC001), National Science Foundation for Young Scientists of China (Nos. 51709271, 41601435).


  1. 1.
    Gevaert CM, Suomalainen J, Tang J, Kooistra L (2015) Generation of spectral–temporal response surfaces by combining multispectral satellite and hyperspectral UAV imagery for precision agriculture applications. IEEE J Sel Top Appl Earth Obs Remote Sens 8(6):3140–3146CrossRefGoogle Scholar
  2. 2.
    Yuen PW, Richardson M (2013) An introduction to hyperspectral imaging and its application for security, surveillance and target acquisition. J Photogr Sci 58(5):241–253Google Scholar
  3. 3.
    Zhang CY, Cheng HF, Chen ZH, Zheng WW (2008) The development of hyperspectral remote sensing and its threatening to military equipments. Electro-Opt Technol Appl 23(1):10–12Google Scholar
  4. 4.
    Sandidge JC, Holyer RJ (1998) Coastal bathymetry from hyperspectral observations of water radiance. Remote Sens Environ 65(3):341–352CrossRefGoogle Scholar
  5. 5.
    Gao HM, Yang Y, Li CM, Zhou H, Qu XY (2018) Joint alternate small convolution and feature reuse for hyperspectral image classification. ISPRS Int J Geo-Inf 7(9):349CrossRefGoogle Scholar
  6. 6.
    Chen Y, Nasrabadi NM, Tran TD (2013) Hyperspectral image classification via kernel sparse representation. IEEE Trans Geosci Remote Sens 51(1):217–231CrossRefGoogle Scholar
  7. 7.
    Qian Y, Ye M, Zhou J (2013) Hyperspectral image classification based on structured sparse logistic regression and three-dimensional wavelet texture features. IEEE Trans Geosci Remote Sens 51(4):2276–2291CrossRefGoogle Scholar
  8. 8.
    Melgani F, Bruzzone L (2004) Classification of hyperspectral remote sensing images with support vector machines. IEEE Trans Geosci Remote Sens 42(8):1778–1790CrossRefGoogle Scholar
  9. 9.
    Li J, Du Q, Li W et al (2015) Optimizing extreme learning machine for hyperspectral image classification. J Appl Remote Sens 9(1):097296CrossRefGoogle Scholar
  10. 10.
    Zhang X, Song Q, Gao Z et al (2017) Spectral–spatial feature learning using cluster-based group sparse coding for hyperspectral image classification. IEEE J Sel Top Appl Earth Obs Remote Sens 9(9):4142–4159CrossRefGoogle Scholar
  11. 11.
    Krizhevsky A, Sutskever I, Hinton GE (2012) ImageNet classification with deep convolutional neural networks. In: International conference on neural information processing systems. Curran Associates Inc., pp 1097–1105Google Scholar
  12. 12.
    Mou L, Ghamisi P, Zhu XX (2017) Deep recurrent neural networks for hyperspectral image classification. IEEE Trans Geosci Remote Sens 55(7):3639–3655CrossRefGoogle Scholar
  13. 13.
    Zhang ZW, Sun Z, Liu JQ, Chen JW, Huo Z, Zhang X (2016) Deep recurrent convolutional neural network: improving performance for speech recognition. arXiv:1611.07174v2
  14. 14.
    Xie S, Girshick R, Dollar P, Tu ZW, He KM (2017) Aggregated residual transformations for deep neural networks. arXiv:1611.05431v2
  15. 15.
    Young T, Hazarika D, Poria S, Cambria E (2018) Recent trends in deep learning based natural language processing. arXiv:1708.02709v6
  16. 16.
    Yu SQ, Jia S, Xu CY (2016) Convolutional neural networks for hyperspectral image classification. Neurocomputing 219:88–98CrossRefGoogle Scholar
  17. 17.
    Ren SQ, He KM, Girshick R, Sun J (2015) Faster R-CNN: towards real-time object detection with region proposal networks. In: International conference on neural information processing systems. MIT Press, pp 91–99Google Scholar
  18. 18.
    Makantasis K, Karantzalos K, Doulamis A, et al (2015) Deep supervised learning for hyperspectral data classification through convolutional neural networks. In: Geoscience and remote sensing symposium (IGARSS), 2015 IEEE international. IEEEGoogle Scholar
  19. 19.
    Gao HM, Yang Y, Lei S, Li CM, Zhou H, Qu XY (2019) Multi-branch fusion network for hyperspectral image classification. Knowl-Based Syst 2019(167):11–25CrossRefGoogle Scholar
  20. 20.
    Chen YS, Zhu L, Ghamisi P, Jia XP, Li GY, Tang L (2017) Hyperspectral images classification with gabor filtering and convolutional neural network. IEEE Geosci Remote Sens Lett 14(12):2355–2359CrossRefGoogle Scholar
  21. 21.
    Liang H, Li Q (2016) Hyperspectral imagery classification using sparse representations of convolutional neural network features. Remote Sens 8(2):99CrossRefGoogle Scholar
  22. 22.
    Lee H, Kwon H (2017) Going deeper with contextual CNN for hyperspectral image classification. IEEE Trans Image Process 26(10):4843–4855MathSciNetCrossRefGoogle Scholar
  23. 23.
    Paoletti ME, Haut JM, Plaza J, Plaza A (2018) A new deep convolutional neural network for fast hyperspectral image classification. ISPRS J Photogramm Remote Sens 145:120–147. CrossRefGoogle Scholar
  24. 24.
    Cao X, Zhou F, Xu L et al (2018) Hyperspectral image classification with Markov random fields and a convolutional neural network. IEEE Trans Image Process 27(5):2354–2367MathSciNetCrossRefGoogle Scholar
  25. 25.
    Zhong Z, Li J, Luo Z et al (2018) Spectral–spatial residual network for hyperspectral image classification: a 3-D deep learning framework. IEEE Trans Geosci Remote Sens 56(2):847–858CrossRefGoogle Scholar
  26. 26.
    Song W, Li S, Fang L et al (2018) Hyperspectral image classification with deep feature fusion network. IEEE Trans Geosci Remote Sens 56(6):3173–3184CrossRefGoogle Scholar
  27. 27.
    Ma Xiaorui Fu, Anyan Wang Jie et al (2018) Hyperspectral image classification based on deep deconvolution network with skip architecture. IEEE Trans Geosci Remote Sens 56(8):4781–4791CrossRefGoogle Scholar
  28. 28.
    Fang B, Li Y, Zhang HK et al (2019) Hyperspectral images classification based on dense convolutional networks with spectral-wise attention mechanism. Remote Sens 11(2):159CrossRefGoogle Scholar
  29. 29.
    He KM, Zhang XY, Ren SQ, Sun J (2015) Delving deep into rectifiers: surpassing human-level performance on ImageNet classification. arXiv:1502.01852v1
  30. 30.
    Nair V, Hinton GE (2010) Rectified linear units improve restricted boltzmann machines. In: International conference on international conference on machine learning. Omnipress, pp 807–814Google Scholar
  31. 31.
    Ioffe S, Szegedy C (2015) Batch normalization: accelerating deep network training by reducing internal covariate shift. arXiv:1502.03167v3
  32. 32.
    Fang L, Liu G, Li S et al (2019) Hyperspectral image classification with squeeze multibias network. IEEE Trans Geosci Remote Sens 57(3):1291–1301CrossRefGoogle Scholar
  33. 33.
    Shu L, Mcisaac K, Osinski GR (2018) Hyperspectral image classification with stacking spectral patches and convolutional neural networks. IEEE Trans Geosci Remote Sens 56(10):5975–5984CrossRefGoogle Scholar
  34. 34.
    Zhi L, Yu XC, Li B et al (2019) A dense convolutional neural network for hyperspectral image classification. Remote Sens Lett 10(1):59–66CrossRefGoogle Scholar
  35. 35.
    Huang G, Liu Z, Maaten LVD, Weinberger KQ (2017) Densely connected convolutional networks. arXiv:1608.06993v3
  36. 36.
    He KM, Zhang XY, Ren SQ, Sun J (2015) Deep residual learning for image recognition. arXiv:1512.03385v1

Copyright information

© Springer-Verlag London Ltd., part of Springer Nature 2019

Authors and Affiliations

  • Hongmin Gao
    • 1
    • 2
  • Yao Yang
    • 1
  • Chenming Li
    • 1
    Email author
  • Xiaoke Zhang
    • 3
  • Jia Zhao
    • 1
  • Dan Yao
    • 1
  1. 1.College of Computer and InformationHohai UniversityNanjingChina
  2. 2.Nantong Ocean and Coastal Engineering Research InstituteHohai UniversityNantongChina
  3. 3.School of Public AdministrationHohai UniversityNanjingChina

Personalised recommendations