Journal of the Korean Institute of Intelligent Systems (한국지능시스템학회논문지)

Korean Institute of Intelligent Systems (한국지능시스템학회)
권호 표지
DOI QR코드

DOI QR Code

An Experimental Study on Smoothness Regularized LDA in Hyperspectral Data Classification

하이퍼스펙트럴 데이터 분류에서의 평탄도 LDA 규칙화 기법의 실험적 분석

  • 박래정 (강릉원주대학교 전자공학과)
  • Received : 2010.04.19
  • Accepted : 2010.08.01
  • Published : 2010.08.25
Download PDF
⟨ Previous Next ⟩

Abstract

High dimensionality and highly correlated features are the major characteristics of hyperspectral data. Linear projections such as LDA and its variants have been used in extracting low-dimensional features from high-dimensional spectral data. Regularization of LDA has been introduced to alleviate the overfitting that often occurs in a small-sized training data set and leads to poor generalization performance. Among them, a smoothness regularized LDA seems to be effective in the feature extraction for hyperspectral data due to its capability of utilizing the high correlatedness. This paper studies the performance of the regularized LDA in hyperspectral data classification experimentally with varying conditions of the training data. In addition, a new dual smoothness regularized LDA is proposed and evaluated that makes use of both the spectral-domain and spatial-domain correlations between neighboring pixels.

고차원 특성과 높은 상관성은 하이퍼스펙트럴 데이터의 주요 특징이다. LDA와 그 변형 선형 투사 방법들이 고차원 스펙트럴 정보로부터 저차원의 특징을 추출하는데 사용되었다. LDA는 학습 데이터가 적은 경우 흔히 발생하는 과대적합으로 인해 일반화 성능이 낮아지는 문제가 발생하는데, 이를 완화하기 위하여 LDA 규칙화(regularization) 방법들이 제시되었다. 그 중, 평탄도(smoothness) 제약에 기반한 LDA 규칙화 기법은 높은 상관성을 갖는 하이퍼스펙트럴 데이터의 특성에 적합한 특징 추출 기법이다. 본 논문에서는 하이퍼스펙트럴 데이터 분류에서 평탄도 제약을 갖는 LDA 규칙화 방법을 소개하고 학습 데이터 조건에 따른 성능을 실험적으로 분석한다. 또한, 분류 성능의 향상을 위한 스펙트럴 정보와 공간적 정보의 상관성을 함께 활용하는 이중 평탄도 LDA 규칙화 기법을 제시한다.

Keywords

References

  1. P. Martinsen, P. Schaarse, and M. Andrews, "A versatile near-infrared imaging spectrometer," J. of Near-Infrared Spectrosc., vol. 7, no. 1, pp. 17-25, 1999. https://doi.org/10.1255/jnirs.230
  2. G. Shaw and D. Manolakis, "Signal processing for hyperspectral image exploitation," IEEE Signal Processing Magazine, vol. 19, no. 1, pp. 12-16, 2002. https://doi.org/10.1109/79.974715
  3. Z. Du, M. K. Jeong, and S. G. Kong, “Band selection of hyperspectral images for automatic detection of poultry skin tumors,” IEEE Trans. Automation Science and Engineering, vol. 4, no. 3, pp. 332-339, 2007. https://doi.org/10.1109/TASE.2006.888048
  4. G. F. Houghes, "On the mean accuracy of statistical pattern recognition," IEEE Trans. Inform. Theory, vol. 14, no. 1, pp. 55-63, 1968. https://doi.org/10.1109/TIT.1968.1054102
  5. K. Fukunga, Introduction to Statistical Pattern Recognition, Academic Press, New York, 1991.
  6. J. H. Friedman, "Regularized discriminant analysis," J. of the American Statistical Association, vol. 84, no. 405, pp. 165-175, 1989. https://doi.org/10.2307/2289860
  7. J. Lu, K. N. Plataniotis, and A. N. Venetsanopoulos, “Regularization studies of linear discriminant analysis in small sample size scenarios with application to face recognition," Pattern Recognition Letters, vol. 26, no. 2, pp. 181-191, 2005. https://doi.org/10.1016/j.patrec.2004.09.014
  8. D.-Q. Dai and P. C. Yuen, "Face recognition by regularized discriminant analysis," IEEE Trans. on SMC-PART B, vol. 37, no. 4, pp. 1080-1085, 2007.
  9. T. Hastie, A. Buja, and R. Tibshirani, "Penalized discriminant analysis", The Annals of Statistics, vol. 23, no. 1, pp. 73-102, 1995. https://doi.org/10.1214/aos/1176324456
  10. D. Cai, X. He, Y. Hu, J. Han, and T. Husang, "Learning a spatially smooth subspace for face recognition", Proc. of IEEE Conf. on Computer Vision and Pattern Recognition, pp. 1-7, 2007.
  11. B. Yu, I. Michael Ostland, P. Gong, and R. Pu, "Penalized discriminant analysis of in situ hyperspectral data for species recognition," IEEE Trans. on Geoscience and Remote Sensing, vol. 37, no. 5, pp. 2569-2577, 1999. https://doi.org/10.1109/36.789651
  12. R. A. Fisher, "The use of multiple measures in taxonmic problems," Ann. Eugenics, vol. 7, pp. 179-188, 1936. https://doi.org/10.1111/j.1469-1809.1936.tb02137.x
  13. P. N. Belhumeur, J. P. Hespanha, and D. J. Kriegman, “Eigenface vs. Fisherface: Recognition using class specific linear projection,” IEEE Trans. Pattern Analysis Machine Intelligence, vol. 19, pp. 711-720, 1997. https://doi.org/10.1109/34.598228
  14. H. Yu and J. Yang, "A direct LDA algorithm for high-dimensional data with application to face recognition," Pattern Recognition, vol. 34, pp. 2067-2070, 2001. https://doi.org/10.1016/S0031-3203(00)00162-X
  15. L. Chen, H. Liao, M. Ko, J. Lin, and G. Yu, "A new LDA-based face recognition system which can solve the small size problem," Pattern Recognition, vol. 33, no. 10, pp. 1713-1726, 2000. https://doi.org/10.1016/S0031-3203(99)00139-9
  16. D. Landgrebe, "On information extraction principles for hyperspectral data: a white paper," Tech. Rep., School Elec. Comput. Eng., Purdue Univ., West Lafayette, IN, 1997.