Effect of Dynamic Mode Decomposition-Based Dimension Reduction Technique on Hyperspectral Image Classification

  • P. Megha
  • V. Sowmya
  • K. P. Soman
Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 490)


Hyperspectral imaging has become an interesting area of research in remote sensing over the past thirty years. But the main hurdles in understanding and analyzing hyperspectral datasets are the high dimension and presence of noisy bands. This work proposes a dynamic mode decomposition (DMD)-based dimension reduction technique for hyperspectral images. The preliminary step is to denoise every band in a hyperspectral image using least square denoising, and the second stage is to apply DMD on hyperspectral images. In the third stage, the denoised and dimension reduced data is given to alternating direction method of multipliers (ADMMs) classifier for validation. The effectiveness of proposed method in selecting most informative bands is compared with standard dimension reduction algorithms like principal component analysis (PCA) and singular value decomposition (SVD) based on classification accuracies and signal-to-noise ratio (SNR). The results illuminate that the proposed DMD-based dimension reduction technique is comparable with the other dimension reduction algorithms in reducing redundancy in band information.


Least square denoising Dimension reduction Principal component analysis Singular value decomposition Dynamic mode decomposition Alternating direction method of multiplier 


  1. 1.
    Aswathy C, Sowmya V, Soman K (2015) ADMM based hyperspectral image classification improved by denoising using Legendre Fenchel transformation. Indian J Sci Technol 8(24):1CrossRefGoogle Scholar
  2. 2.
    Bhushan DB, Sowmya V, Manikandan MS, Soman K (2011) An effective pre-processing algorithm for detecting noisy spectral bands in hyperspectral imagery. In: 2011 International symposium on ocean electronics, pp 34–39Google Scholar
  3. 3.
    Chen S, Donoho D. Basis pursuit, vol 1, pp 41–44Google Scholar
  4. 4.
    Erichson NB, Brunton SL, Kutz JN (2015) Compressed dynamic mode decomposition for real-time object detection. arXiv preprint arXiv:1512.04205
  5. 5.
    Fong M (2007) Dimension reduction on hyperspectral images. Univ. California, Los Angeles, CAGoogle Scholar
  6. 6.
    Koonsanit K, Jaruskulchai C, Eiumnoh A (2012) Band selection for dimension reduction in hyper spectral image using integrated information gain and principal components analysis technique. Int J Mach Learn Comput 2(3):248CrossRefGoogle Scholar
  7. 7.
    Li Y (2014) Dimension reduction for hyperspectral imaging using laplacian eigenmaps and randomized principal component analysis: midyear re-portGoogle Scholar
  8. 8.
    Lodha SP, Kamlapur S (2014) Dimensionality reduction techniques for hyperspectral images. Int J Appl Innov Eng Manag (IJAIEM) 3(10)Google Scholar
  9. 9.
    Reshma R, Sowmya V, Soman K (2016) Dimensionality reduction using band selection technique for kernel based hyperspectral image classification. Procedia Comput Sci 93:396–402CrossRefGoogle Scholar
  10. 10.
    Rodarmel C, Shan J (2002) Principal component analysis for hyper-spectral image classification. Surv Land Inf Sci 62(2):115Google Scholar
  11. 11.
    Schmid PJ (2010) Dynamic mode decomposition of numerical and experimental data. J Fluid Mech 656:5–28MathSciNetCrossRefzbMATHGoogle Scholar
  12. 12.
    Selesnick I (2013) Least squares with examples in signal processing. Connexions 4Google Scholar
  13. 13.
    Srivatsa S, Ajay A, Chandni C, Sowmya V, Soman K (2016) Application of least square denoising to improve ADMM based hyperspectral image classification. Procedia Comput Sci 93:416–423CrossRefGoogle Scholar
  14. 14.
    Xu L, Li F, Wong A, Clausi DA (2015) Hyperspectral image denoising using a spatial–spectral monte carlo sampling approach. IEEE J Sel Top Appl Earth Obs Remote Sens 8(6):3025–3038CrossRefGoogle Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2018

Authors and Affiliations

  1. 1.Centre for Computational Engineering and Networking (CEN), Amrita School of EngineeringAmrita Vishwa VidyapeethamCoimbatoreIndia

Personalised recommendations