The Journal of Korean Institute of Communications and Information Sciences (한국통신학회논문지)

The Korean Institute of Commucations and Information Sciences (한국통신학회)
권호 표지

MRS Pattern Classification Using Fusion Method based on SpPCA and MLP

SpPCA와 MLP에 기반을 둔 응합법칙에 의한 MRS 패턴분류

  • 송창규 (충북대학교 전기전자컴퓨터공학부) ;
  • 이대종 (충북대학교 전기전자컴퓨터공학부) ;
  • 전병석 ((주)세화폴리텍) ;
  • 유정웅 (충북대학교 전기전자컴퓨터공학부)
  • Published : 2005.09.01
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, we propose the MRS p:Ittern classification techniques by the fusion scheme based on the SpPCA and MLP. A conventional PCA teclulique for the dimension reduction has the problem that it can't find a optimal transformation matrix if the property of input data is nonlinear. To overcome this drawback we extract features by the SpPCA technique which use the local patterns rather than whole patterns. In a next classification step, individual classifier based on MLP calculates the similarity of each class for local features. Finally, MRS patterns is classified by the fusion scheme to effectively combine the individual information. As the simulation results to verify the effectiveness, the proposed method showed more improved classification results than conventional methods.

본 논문에서는 SpPCA와 MLP에 기반을 둔 융합법칙에 의한 MRS 패턴분류기법을 제안한다. 차원축소를 위해 사용되는 기존의 PCA 기법은 입력데이터가 비선형 특성을 갖는 경우 최적의 변환행렬을 구할 수 없다는 문제점을 가지고 있다. 따라서, 본 논문에서는 구간별로 입력데이터를 분할한 후 PCA에 의해 특징을 추출하는 SpPCA 기법을 이용하여 입력패턴의 차원을 축소한다. 다음 단계인 분류단계에서는 MLP 비선형분류기를 이용하여 구간마다 추출된 특징벡터를 이용하여 기준패턴과의 유사도를 산출한다. 최종 분류단계에서는 MLP에 의해서 산출된 유사도에 기반을 둔 융합법칙에 의하여 MRS 패턴을 분류한다. 제안된 방법의 유용성을 보이기 위한 실험결과에서 기존의 방법들에 비해서 향상된 인식결과를 보임을 확인하였다.

Keywords

References

  1. R. Baumgarttner, R. Somorjai, C. Bowman, T.C. Sorrell, C.E. Mountford, U. Himmelreich, Unsupervised feature dimension reduction for classification of MR spectra, Magnetic Resonance Imaging 22, pp. 251-256, 2004 https://doi.org/10.1016/j.mri.2003.08.033
  2. Y.Y.B. Lee, Y. Huang, W. El-Deredy, P.J.G. Lisboa, C. Arus, P. Harris, Robust methodology for the discrimination of brain tumours from in vivo magnetic resonance spectra: Science, Measurement and Technology, IEE Proc., Vol.147, Issue 6, pp.309-314, 2000 https://doi.org/10.1049/ip-smt:20000850
  3. Yu Wang, S. Van Huffel, E. Heyvaert, L. Vanhamme, N. Mastronardi, P. Van Hecke, Magnetic resonance spectroscopic quantitation via complex principal component analysis, WCCC-ICSP 2000, Vol.3, pp.2074-2077, 2000
  4. M. Turk and A. Pentland, Face recognition using eigenfaces, IEEE Conf. on Computer Vision and Pattern Recognition, pp.586-591, 1991
  5. Hong Yang, Joseph Irudayaraj and Manish M. Paradkar, Discriminant analysis of edible oils and fats by FTIR, FT-NIR and FT-Raman spectroscopy, Food Chemistry, Vol.93, Issue 1, pp.25-32, 2005 https://doi.org/10.1016/j.foodchem.2004.08.039
  6. Marina Vannucci, Naijun Sha and Philip J. Brown, NIR and mass spectra classification: Bayesian methods for wavelet-based feature selection, Chemometrics and Intelligent Laboratory Systems, In Press, Corrected Proof, Available online 4 March 2005
  7. E.K. Tang, P.N. Suganthan, X. Yao and A.K. Qin, Linear dimensionality reduction using relevance weighted LDA, Pattern Recognition, Vol.38, Issue4, pp.485-493, 2005 https://doi.org/10.1016/j.patcog.2004.09.005
  8. N.P. Hughes, S.J. Roberts, L. Tarassenko, Semi-supervised learning of probabilistic models for ECG segmentation, Engineering in Medicine and Biology Society, EMBC 2004, Vol.1, pp.434-437, 2004
  9. Songcan Chen, Yulian Zhu, Subpattern- based principal component analysis, Pattern Recognition 37, pp.1081-1083, 2004 https://doi.org/10.1016/j.patcog.2003.09.004
  10. Rajkiran Gottumukkal, Vijayan K. Asari, An improved face recognition technique based on modular PCA approach, Pattern Recognition Letters 25, pp.429-436, 2004 https://doi.org/10.1016/j.patrec.2003.11.005
  11. Yang-Lang Chang, Chin-Chuan Han, Fan-Di Jou, Kuo-Chin Fan, K.S. Chen, Jeng-Horng Chang, A modular eigen subspace scheme for high-dimensional data classification, Future Generation Computer System 20, pp.1131-1143, 2004 https://doi.org/10.1016/j.future.2003.11.003
  12. Nicolino J. Pizzi, Fuzzy pre-processing of gold standards as applied to biomedical spectra classification, Artificial Intelligence in Medicine