Journal of KIISE:Software and Applications (한국정보과학회논문지:소프트웨어및응용)

Korean Institute of Information Scientists and Engineers (한국정보과학회)
권호 표지

Image Classification Method using Independent Component Analysis and Normalization

독립성분해석과 정규화를 이용한 영상분류 방법

  • Published : 2001.09.01
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, we improve noise tolerance in image classification by combining ICA(Independent Component Analysis) with Normalization. When we add noise to the raw image data the degree of noise tolerance becomes N(0, 0.4) for PCA and N(0, 0.53) for ICA. However, when we use the preprocessing approach the degree of noise tolerance after Normalization becomes N(0, 0.75), which shows the improvement of noise tolerance in classification.

본 논문에서는 독립 성분 해석(Independent Component Analysis, ICA) 기법과 정규화를 이용한 영상분류 방법을 제안한다. 이 제안된 방법은 전처리 없이 ICA나 주성분 해석(Principal Component Analysis, PCA)을 이용한 것에 비해 잡음에 대한 강인성을 증가시킨다. 영상에 잡음이 인가된 경우, CPA는 N(0, 0.4), ICA는 N(0.53)까지이 분류가 가능함을 보이는 반면에 비해, 제안된 정규화 전처리는 N(0, 0.75)까지 영상분류가 됨을 실험에서 보이고 있다.

Keywords

References

  1. Comon, P., Independent Component Analysis-a new concept?, signal processing, Vol. 36, pp.287-314, 1994 https://doi.org/10.1016/0165-1684(94)90029-9
  2. Oja, E., The nonlinear PCA learning rule in independent component analysis, Neurocomputing, Vol. 17, No. 1, pp.25-46, 1997 https://doi.org/10.1016/S0925-2312(97)00045-3
  3. Pajunen, P., Blind source separation using algorithmic information theory., Neurocomputing, 1998 https://doi.org/10.1016/S0925-2312(98)00048-4
  4. Delfosse, N. and Loubaton, P., Adaptive blind separation of independent sources : a deflation approach., Signal Processing, Vol. 45, pp.59-83, 1995 https://doi.org/10.1016/0165-1684(95)00042-C
  5. Hyvarinen, A. and Oja., E., A fast fixed-point algorithm for independent component analysis, Neural Computation, Vol. 9, No. 7, pp.1483-1492, 1997 https://doi.org/10.1162/neco.1997.9.7.1483
  6. Parkkinen, J., Jaaskelainen, T., Color Representation Using Statistical Pattern Recognition, Applied Optics, Vol. 26, No. 19, pp.4240-4245, 1987