Multimedia Tools and Applications

, Volume 78, Issue 3, pp 3411–3433 | Cite as

DCNR: deep cube CNN with random forest for hyperspectral image classification

  • Tao Li
  • Jiabing Leng
  • Lingyan Kong
  • Song Guo
  • Gang Bai
  • Kai WangEmail author


Hyperspectral Image (HSI) classification is one of the fundamental tasks in the field of remote sensing data analysis. CNN (Convolutional Neural Network) has been proven to be an effective deep learning model, which can extract high-level features directly from the raw data and thereby utilize rich information contained in HSI data. However, labor cost to label enough HIS data for training model is usually expensive, so that it is a strong demand of utilizing limited training data to get a satisfied classification accuracy. In this paper, we put forward a deep cube CNN model – DCNR, which is composed of a cube neighbor HSI pixels strategy, a deep CNN and a random forest classifier. In DCNR model, cubic samples, containing spectral-spatial information, are generated by putting each target pixel and its neighbors together. Then features with high representative ability, extracted by applying a specially designed cube CNN model on each cubic sample, are fed into the random forest classifier for the classification of the target pixel. Results show that DCNR model can achieve classification accuracy of 96.78%, 96.08% and 94.85% on KSC, IP and SA datasets respectively with 20% samples as training set, and 85.03%, 83.45 and 62.17% on KSC, IP and SA datasets respectively with only 1% samples as training set, significantly outperforming random forest and cube CNN models.


HSI classification Deep learning CNN Random forest Spectral-spatial feature 



This study was funded by the Natural Science Foundation of Tianjin under Grant No. 16JCYBJC15200, the Major Science and Technology Program of Big Data and Cloud Computing of Tianjin No. 15ZXDSGX00020, the Science and Technology Commission of Tianjin Binhai New Area No. BHXQKJXM-PT-ZJSHJ-2017005, the National Key Research and Development Program of China (2016YFC0400709), and the Fundamental Research Funds for the Central Universities.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.


  1. 1.
    Alsmirat MA, Jararweh Y, Al-Ayyoub M et al (2017) Accelerating compute intensive medical imaging segmentation algorithms using hybrid CPU-GPU implementations[J]. Multimed Tools Appl 76(3):3537–3555CrossRefGoogle Scholar
  2. 2.
    AlZain MA, Li AS, Soh B, Pardede E (2015) Multi-cloud data management using Shamir's secret sharing and quantum byzantine agreement schemes[J]. Int J Cloud Appl Comput (IJCAC) 5(3):35–52Google Scholar
  3. 3.
    Atawneh S, Almomani A, Al Bazar H et al (2017) Secure and imperceptible digital image steganographic algorithm based on diamond encoding in DWT domain[J]. Multimed Tools Appl 76(18):18451–18472CrossRefGoogle Scholar
  4. 4.
    Bergstra J, Breuleux O, Bastien F, et al. (2010) Theano: a CPU and GPU math compiler in Python[C]//proc. 9th Python in Science Conf : 1–7Google Scholar
  5. 5.
    Bioucas-Dias JM, Plaza A, Camps-Valls G et al (2013) Hyperspectral remote sensing data analysis and future challenges[J]. Geosci Remote Sens Mag, IEEE 1(2):6–36CrossRefGoogle Scholar
  6. 6.
    Camps-Valls G, Bruzzone L (2005) Kernel-based methods for hyperspectral image classification[J]. IEEE Trans Geosci Remote Sens 43(6):1351–1362CrossRefGoogle Scholar
  7. 7.
    Camps-Valls G, Tuia D, Bruzzone L et al (2013) Advances in hyperspectral image classification: earth monitoring with statistical learning methods[J]. IEEE Signal Process Mag 31(1):45–54CrossRefGoogle Scholar
  8. 8.
    Chan W, Jaitly N, Le Q et al. (2016) Listen, attend and spell: a neural network for large vocabulary conversational speech recognition[C]//acoustics, speech and signal processing (ICASSP). 2016 I.E. Int Conf. IEEE: 4960–4964Google Scholar
  9. 9.
    Chang X, Ma Z, Lin M, Yang Y, Hauptmann AG (2017) Feature interaction augmented sparse learning for fast kinect motion detection[J]. IEEE Trans Image Process 26(8):3911–3920MathSciNetCrossRefGoogle Scholar
  10. 10.
    Chang X, Ma Z, Yang Y, Zeng Z, Hauptmann AG (2017) Bi-level semantic representation analysis for multimedia event detection[J]. IEEE transactions on cybernetics 47(5):1180–1197CrossRefGoogle Scholar
  11. 11.
    Chang X, Yang Y (2017) Semisupervised feature analysis by mining correlations among multiple tasks[J]. IEEE Trans Neural Netwrks Learn Syst 28(10):2294–2305MathSciNetCrossRefGoogle Scholar
  12. 12.
    Chang X, Yu YL, Yang Y, Xing EP (2017) Semantic pooling for complex event analysis in untrimmed videos[J]. IEEE Trans Pattern Anal Mach Intell 39(8):1617–1632CrossRefGoogle Scholar
  13. 13.
    Chen Y, Lin Z, Zhao X, Wang G, Gu Y (2014) Deep learning-based classification of hyperspectral data[J]. IEEE J Select Top Appl Earth Observ Remote Sens 7(6):2094–2107CrossRefGoogle Scholar
  14. 14.
    Chen Y, Zhao X, Jia X (2015) Spectral–spatial classification of hyperspectral data based on deep belief network[J]. IEEE J Select Top Appl Earth Observ Remote Sens 8(6):2381–2392CrossRefGoogle Scholar
  15. 15.
    Gu J, Wang Z, Kuen J, et al. (2015) Recent advances in convolutional neural networks[J]. Comput SciGoogle Scholar
  16. 16.
    Gupta S, Gupta BB (2018) XSS-secure as a service for the platforms of online social network-based multimedia web applications in cloud[J]. Multimed Tools Appl 77(4):4829–4861CrossRefGoogle Scholar
  17. 17.
    Ham J, Chen Y, Crawford MM et al (2005) Investigation of the random forest framework for classification of hyperspectral data[J]. IEEE Trans Geosci Remote Sens 43(3):492–501CrossRefGoogle Scholar
  18. 18.
    Harsanyi JC, Chang CI (1994) Hyperspectral image classification and dimensionality reduction: an orthogonal subspace projection approach[J]. IEEE Trans Geosci Remote Sens 32(4):779–785CrossRefGoogle Scholar
  19. 19.
    Heaton JB, Polson NG, Witte JH (2017) Deep learning for finance: deep portfolios[J]. Appl Stoch Model Bus Ind 33(1):3–12MathSciNetCrossRefGoogle Scholar
  20. 20.
    Hu W, Huang Y, Wei L et al. (2015) Deep convolutional neural networks for hyperspectral image classification[J]. J Sens: 2015Google Scholar
  21. 21.
    IGI Global (2016) Handbook of research on modern cryptographic solutions for computer and cyber security[M]Google Scholar
  22. 22.
    Jararweh Y, Al-Ayyoub M, Fakirah M, et al. (2017) Improving the performance of the needleman-wunsch algorithm using parallelization and vectorization techniques[J]. Multimed Tools Appl: 1–17Google Scholar
  23. 23.
    Kang X, Li S, Benediktsson JA (2014) Spectral–spatial hyperspectral image classification with edge-preserving filtering[J]. IEEE Trans Geosci Remote Sens 52(5):2666–2677CrossRefGoogle Scholar
  24. 24.
    Kim Y (2014) Convolutional neural networks for sentence classification[J]. arXiv preprint arXiv:1408.5882Google Scholar
  25. 25.
    Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks[C]//. Adv Neural Inf Proces Syst: 1097–1105Google Scholar
  26. 26.
    LeCun Y, Bengio Y, Hinton G (2015) Deep learning[J]. Nature 521(7553):436–444CrossRefGoogle Scholar
  27. 27.
    Leng J, Li T, Bai G, et al. (2016) Cube-CNN-SVM: a novel hyperspectral image classification method[C]//tools with artificial intelligence (ICTAI). 2016 I.E. 28th Int Conf. IEEE: 1027–1034Google Scholar
  28. 28.
    Li J, Huang X, Gamba P, Bioucas-Dias JMB, Zhang L, Benediktsson JA, Plaza A (2015) Multiple feature learning for hyperspectral image classification[J]. IEEE Trans Geosci Remote Sens 53(3):1592–1606CrossRefGoogle Scholar
  29. 29.
    Li Y, Peng Z, Liang D, Chang H, Cai Z (2016) Facial age estimation by using stacked feature composition and selection. Vis Comput 32(12):1525–1536. CrossRefGoogle Scholar
  30. 30.
    Li W, Prasad S, Fowler JE, Bruce LM (2011) Locality-preserving discriminant analysis in kernel-induced feature spaces for hyperspectral image classification[J]. IEEE Geosci Remote Sens Lett 8(5):894–898CrossRefGoogle Scholar
  31. 31.
    Li Y, Wang G, Nie L, Wang Q, Tan W (2018) Distance metric optimization driven convolutional neural network for age invariant face recognition[J]. Pattern Recogn 75:51–62. CrossRefGoogle Scholar
  32. 32.
    Liaw A, Wiener M (2002) Classification and regression by random forest[J]. R News 2(3):18–22Google Scholar
  33. 33.
    Litjens G, Kooi T, Bejnordi BE, Setio AAA, Ciompi F, Ghafoorian M, van der Laak JAWM, van Ginneken B, Sánchez CI (2017) A survey on deep learning in medical image analysis[J]. Med Image Anal 42:60–88CrossRefGoogle Scholar
  34. 34.
    Maggiori E, Tarabalka Y, Charpiat G, Alliez P (2017) Convolutional neural networks for large-scale remote-sensing image classification[J]. IEEE Trans Geosci Remote Sens 55(2):645–657CrossRefGoogle Scholar
  35. 35.
    Mairal J, Bach F, Ponce J et al. (2009) Online dictionary learning for sparse coding[C]//. Int Conf Mach Learn, ICML 2009, Montreal, Quebec, Canada, June. DBLP:689–696Google Scholar
  36. 36.
    Majumder N, Poria S, Gelbukh A, Cambria E (2017) Deep learning-based document modeling for personality detection from text[J]. IEEE Intell Syst 32(2):74–79CrossRefGoogle Scholar
  37. 37.
    Nair V, Hinton G E. (2010) Rectified linear units improve restricted Boltzmann machines[C]//. Int Conf Mach Learn. DBLP, 807–814Google Scholar
  38. 38.
    Qi CR, Su H, Mo K et al (2017) Pointnet: deep learning on point sets for 3d classification and segmentation[J]. Proc Comput Vision Pattern Recogn (CVPR), IEEE 1(2):4Google Scholar
  39. 39.
    Quang D, Xie X, Dan Q (2016) A hybrid convolutional and recurrent deep neural network for quantifying the function of DNA sequences[J]. Nucleic Acids Res 44(11):e107CrossRefGoogle Scholar
  40. 40.
    Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition[J]. arXiv preprint arXiv:1409.1556,Google Scholar
  41. 41.
    Vincent P, Larochelle H, Bengio Y et al. (2008) Extracting and composing robust features with denoising autoencoders[C]//. Int Conf :1096–1103Google Scholar
  42. 42.
    Vincent P, Larochelle H, Lajoie I et al (2010) Stacked denoising autoencoders: learning useful representations in a deep network with a local denoising criterion.[J]. J Mach Learn Res 11(12):3371–3408MathSciNetzbMATHGoogle Scholar
  43. 43.
    Wang K, Zhang D, Ya L, Zhang R, Lin L (2017) Cost-effective active learning for deep image classification. IEEE Trans Circ Syst Video Technol (T-CSVT) 27(12):2591–2600CrossRefGoogle Scholar
  44. 44.
    Xia J, Du P, He X et al (2014) Hyperspectral remote sensing image classification based on rotation forest[J]. IEEE Geosci Remote Sens Lett 11(1):239–243CrossRefGoogle Scholar
  45. 45.
    Xie L, Li G, Xiao M, Peng L, Chen Q (2017) Hyperspectral image classification using discrete space model and support vector machines[J]. IEEE Geosci Remote Sens Lett 14(3):374–378CrossRefGoogle Scholar
  46. 46.
    Yuan C, Li X, Wu QMJ et al (2017) Fingerprint liveness detection from different fingerprint materials using convolutional neural network and principal component analysis[J].CMC: computers. Mater Continua 53(3):357–371. CrossRefGoogle Scholar
  47. 47.
    Zeng S, Bai J, Jiang L, et al. (2017) Multiple kernel fuzzy discriminant analysis for hyperspectral imaging classification[C]//fuzzy systems (FUZZ-IEEE). 2017 I.E. Int Conf IEEE: 1–6Google Scholar
  48. 48.
    Zhang L, Zhang L, Tao D, Huang X (2012) On combining multiple features for hyperspectral remote sensing image classification[J]. IEEE Trans Geosci Remote Sens 50(3):879–893CrossRefGoogle Scholar
  49. 49.
    Zhao W, Du S (2016) Spectral–spatial feature extraction for hyperspectral image classification: a dimension reduction and deep learning approach[J]. IEEE Trans Geosci Remote Sens 54(8):4544–4554CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  1. 1.College of Computer and Control Engineering, Sino-Canada Joint R&D Centre on Water and Environmental SafetyNankai UniversityTianjinPeople’s Republic of China

Personalised recommendations