Communications for Statistical Applications and Methods

The Korean Statistical Society (한국통계학회)
권호 표지

Feature Extraction and Statistical Pattern Recognition for Image Data using Wavelet Decomposition

  • Kim, Min-Soo (Department of Statistics, Chonnam National University) ;
  • Baek, Jang-Sun (Department of Statistics, Chonnam National University)
  • Published : 1999.12.01
Download PDF
⟨ Previous Next ⟩

Abstract

We propose a wavelet decomposition feature extraction method for the hand-written character recognition. Comparing the recognition rates of which methods with original image features and with selected features by the wavelet decomposition we study the characteristics of the proposed method. LDA(Linear Discriminant Analysis) QDA(Quadratic Discriminant Analysis) RDA(Regularized Discriminant Analysis) and NN(Neural network) are used for the calculation of recognition rates. 6000 hand-written numerals from CENPARMI at Concordia University are used for the experiment. We found that the set of significantly selected wavelet decomposed features generates higher recognition rate than the original image features.

Keywords

References

  1. 제2회 문자인식워크샵 발표논문집 통계적 군집화 및 신경망기법을 이용한 필기체 문자의 오프라인 인식 김수형(외 3인)
  2. 한국 정보과학회 논문지 v.19 no.2 다양한 활자체 및 크기의 한글 문자 영상에서의 정보량 및 엔트로피의 분포 이성환
  3. 문자인식 - 이론과 실제 이성환
  4. 패턴인식의 원리 이성환
  5. S+Wavelets - User's Manual Bruce, A.;Gao, H.
  6. Improved Linear Discriminant Using Time-Frequency Dictionaries Buckheit, J.;Donoho, D.
  7. IEEE Trans. Image Process v.5 Wavelet descriptor of planar curves: theory and applications Chang, G.C.H.;Jay Kuo, C.C.
  8. Comptes. Rendus. Acad. Sci. Paris v.Serie I 319 no.2 Constructions of local orthonormal bases for classification and regression Coifman, R.R.;Saito, N.
  9. IEEE Trans. Inform. Theory v.38 no.2 Entropy-based algorithms best basis selection Coifman, R.R.;Wickerhauser, M.V.
  10. Journal of the American Statistical Association v.84 Regularized discriminant analysis Friedman, J.H.
  11. Proc. 6th IWFHR Performance comparison of statistical and neural network classifiers in hand-written digits recognition Jeong, S.W.;Kim, S.H.;Cho, W.H.
  12. Essential Wavelets for Statistical Applications and Data Analysis Ogden, R.T.
  13. Bell System Technical Journal v.27 A Mathematical Theory of Communications Shannon, C.E.
  14. ICPR'96 v.II A novel approach to optical character recognition based on ring-projection-wavelet-fractal signatures Tang, Y.Y.;Li, B.F.;Ma, H.;Liu, J.;Leung, C.H.;Suen, C.Y.
  15. Information Science v.27 no.1 A general scheme for constructing skeleton models Tsao, Y.F.;Fu, K.S.
  16. Pattern Recognition v.27 Invariant pattern recognition by moment Fourier descriptor Wang, S.S.;Chen, P.C.;Lin, W.G.
  17. Pattern Recognition v.28 Wavelet descriptors for multiresolution recognition of handprinted characters Wunsch, P.;Laine, A.F.
  18. Pattern Recognition v.23 A Linear Normalization for Hand-printed Kanji Character Recognition - Line Density Equalization Yamada, H.;Yamamoto, K.;Saito, T.
  19. Proc. 9th Int. Conf. on Pattern Recognition A maximum algorithm for thinning digital pattern Zhang, T.Y.;Wang, P.S.P