이종 특징 결합을 이용한 하이브리드 패턴 인식 방법 = Hybrid Pattern Recognition Using a Combination of Different Features

다국어 초록 (Multilingual Abstract)

We propose a hybrid pattern recognition method that effectively combines two different features for improving data classification. We first extract the PCA (Principal Component Analysis) and LDA (Linear Discriminant Analysis) features, both of which are widely used in pattern recognition, to construct a set of basic features, and then evaluate the separability of each basic feature. According to the results of evaluation, we select only the basic features that contain a large amount of discriminative information for construction of the combined features. The experimental results for the various data sets in the UCI machine learning repository show that using the proposed combined features give better recognition rates than when solely using the PCA or LDA features.

목차 (Table of Contents)

• Abstract
• Ⅰ. Introduction
• Ⅱ. Feature Extraction Methods
• Ⅲ. Hybrid Pattern Classification
• Ⅳ. Experimental Results
• Ⅴ. Conclusions
• REFERENCES

참고문헌 (Reference)

1 허경용, "학습 샘플 선택을 이용한 교사 랭크 정규화" 한국컴퓨터정보학회 20 (20): 21-28, 2015

2 정성태, "조명 변화에 견고한 얼굴 특징 추출" 한국컴퓨터정보학회 10 (10): 1-8, 2005

3 허경용, "의사 샘플 신경망에서 학습 샘플 및 특징 선택 기법" 한국컴퓨터정보학회 18 (18): 19-26, 2013

4 J. Yang, "Two-dimensional PCA: a new approach to appearance-based face representation and recognition" 26 (26): 131-137, 2004

5 M. R. Sikonja, "Theoretical and Empirical Analysis of ReliefF and RReliefF" 53 (53): 23-69, 2003

6 "Repository of Machine Learning Databases"

7 S.-I. Choi, "Pixel selection based on discriminant features with application to face recognition" 33 (33): 1083-1092, 2012

8 A.M. Martínez, "PCA versus LDA" 23 (23): 228-233, 2001

9 I. Kononenko, "Overcoming the myopia of inductive learning algorithms with RELIEFF" 7 (7): 39-55, 1997

10 J. Kittler, "On combining classifiers" 20 (20): 226-239, 1998

11 J. Liang, "Invariant optimal feature selection: a distance discriminant and feature ranking based solution" 41 (41): 1429-1439, 2008

12 S.-I. Choi, "Input variable selection for feature extraction in classification problems" 92 (92): 636-648, 2011

13 N. Kwak, "Input feature selection for classification problems" 13 (13): 143-159, 2002

14 N. Kwak, "Input feature selection by mutual information based on parzen window" 24 (24): 1667-1671, 2002

15 J. Ravi, "Hybrid domain based face recognition system" 3 (3): 1402-1408, 2012

16 T. M. Cover, "Geometrical and Statistical properties of systems of linear inequalities with applications in pattern recognition" EC-14 (EC-14): 326-334, 1965

17 N. Kwak, "Feature selection and extraction based on mutual information for classification" Seoul National University 2003

18 N. Kwak, "Feature extraction for one-class classification problems: Enhancements to biased discriminant analysis" 42 (42): 17-26, 2009

19 S.I. Choi, "Face Recognition Based on 2D Images Under Various Conditions" Seoul National University 2010

20 P.N. Belhumeur, "Eigenfaces vs. fisherfaces: recognition using class specific linear projection" 19 (19): 711-720, 1997

21 M. Turk, "Eigenfaces for recognition" 3 (3): 71-86, 1991

22 H. Cevikalp, "Discriminative common vectors for face recognition" 27 (27): 4-13, 2005

23 C. Kim, "Combined subspace method using global and local features for face recognition" 2030-2035, 2005

24 J. Doak, "An evaluation of feature selection methods and their application to computer security" University of California 1992

25 X. He, "Advances in Neural Information Processing Systems" 507-514, 2005

26 S.I. Choi, "A discriminant distance based composite vector selection method for odor classification" 14 (14): 6938-6951, 2014

27 C. Kim, "A discriminant analysis using composite features for classification problems" 40 (40): 2118-2125, 2007

28 H. Yu, "A direct LDA algorithm for high-dimensional data — with application to face recognition" 34 (34): 2067-2070, 2001

29 C. Kim, "A New Biased Discriminant Analysis Using Composite Vectors for Eye Detection" 42 (42): 1095-1106, 2012