Journal of the Institute of Electronics Engineers of Korea CI (전자공학회논문지CI)

The Institute of Electronics and Information Engineers (대한전자공학회)
권호 표지

Face Recognition based on Hybrid Classifiers with Virtual Samples

가상 데이터와 융합 분류기에 기반한 얼굴인식

  • Published : 2003.01.01
Download PDF
⟨ Previous Next ⟩

Abstract

This paper presents a novel hybrid classifier for face recognition with artificially generated virtual training samples. We utilize both the nearest neighbor approach in feature angle space and a connectionist model to obtain a synergy effect by combining the results of two heterogeneous classifiers. First, a classifier called the nearest feature angle (NFA), based on angular information, finds the most similar feature to the query from a given training set. Second, a classifier has been developed based on the recall of stored frontal projection of the query feature. It uses a frontal recall network (FRN) that finds the most similar frontal one among the stored frontal feature set. For FRN, we used an ensemble neural network consisting of multiple multiplayer perceptrons (MLPs), each of which is trained independently to enhance generalization capability. Further, both classifiers used the virtual training set generated adaptively, according to the spatial distribution of each person's training samples. Finally, the results of the two classifiers are combined to comprise the best matching class, and a corresponding similarit measure is used to make the final decision. The proposed classifier achieved an average classification rate of 96.33% against a large group of different test sets of images, and its average error rate is 61.5% that of the nearest feature line (NFL) method, and achieves a more robust classification performance.

본 논문은 인위적으로 생성된 가상 학습 데이터와 융합 분류기를 이용한 얼굴인식 알고리즘을 제안한다. 특징공간에서의 최근접 특징 선택 방법과 연결주의 모델에 기반한 서로 다른 형태의 분류기를 융합하여 통합효과를 얻도록 하였다. 두 분류기는 모두 학습 데이터의 공간적인 분포에 따라 생성된 가상 학습데이터를 이용하여 학습되고 이용된다. 첫째로, 특징 공간에서의 각 정보(Angular Infnrmation) 를 이용하는 최근접특징각(the Nearest Feature Angle : NFA)을 이용하여 저장된 학습데이터와 가장 근접한 것을 찾고, 둘째로, 질의(Query) 얼굴 특징 정보를 정면얼굴 영상의 특징정보로 투영하여 얻은 정보에 기반한 분류기의 결과를 이용한다. 정면영상 특징정보로의 투영은 다층 신경망을 이용하여 정면 회상망(Frontal Recall Network)을 구현하였고, 이것을 여러 개 묶어 앙상블 네트웍으로 구성한 Ensemble 회상망(Ensemble Recall Network)을 사용하여 일반화 성능을 향상시켰다. 끝으로, 각 분류기의 결과에 따라 융합 분류기가 최종 결과를 선택하도록 하였다. 제안된 알고리즘을 6 종류의 서고 다른 학습/시험데이터 군에 적용하여 평균 96.33%의 인식률을 얻었다. 이것은 특징라인에 기반한 방법(the Nearest Feature Line) 평균 에러율의 61.2% 이며, 단일 분류기를 사용한 경우 보다 안정된 견과를 얻고 있다.

Keywords

References

  1. Stan Z. Li, Juwei Lu, Face Recognition Using the Nearest Feature Line Method, IEEE Transactions on Neural Networks, 10(2), pp. 439-443, 1999 https://doi.org/10.1109/72.750575
  2. Y. Moses, Y. Adini, S. Ullman, Face recognition : The problem of compensating for changes in illumination direction, Proc. European Conf. Computer Vision, vol. A, pp. 286-296, 1994 https://doi.org/10.1007/3-540-57956-7_33
  3. A. Samal, P. A. Iyengar, Automatic recognition and analysis of human faces and facial expression: A survey, Pattern Recognition, 25, pp. 65-77, 1992 https://doi.org/10.1016/0031-3203(92)90007-6
  4. R. Brunelli, T. Poggio, Face Recognition: Features versus templates, IEEE Trans. Pattern Anal. Machine Intelligence, 15, pp. 1042-1052, 1993 https://doi.org/10.1109/34.254061
  5. M. A. Turk, A. P. pentland, Eigenfaces for recognition, J. Cognitive Neuroscience, 4(1), pp. 71-86, 1991
  6. A. J. O'Toole, H. Abdi, K. A. Deffenbacher, D. Valentin, Low dimensional representation of faces in higher dimensions of the face space, J. Opt. Soc. Amer. A(10), pp. 405-410, 1993 https://doi.org/10.1364/JOSAA.10.000405
  7. Jun Zhang, Yong Yan, Martin Lades, Face Recognition: Eigenface, Elastic Matching, and Neural Nets, Proceeding of the IEEE, 85(9), pp. 1423-1435, 1997 https://doi.org/10.1109/5.628712
  8. in Jang, Sungzoon Cho, Ensemble Learning Using Observational Learning Theory, International Joint Conference on Neural Network, 1999
  9. S. Cho, K. Cha, Evolution of neural network training set through addition of virtual samples, International Conference on Evolutionary Computation, Nagoya, Japan, pp. 685-688, 1996 https://doi.org/10.1109/ICEC.1996.542684
  10. L. Holmstrom, P. Koistinen, Using additive noise in back-propagation training, IEEE transactions on Neural Networks, 3. pp. 24-38, 1992 https://doi.org/10.1109/72.105415
  11. C. M. Bishop, Training with noise is equivalent to tikhonov regularization, Neural Computation, 7(1), pp. 108-116, 1995 https://doi.org/10.1162/neco.1995.7.1.108
  12. Y. Abu-Mostafa, Hints, Neural Computation, 7, pp. 639-671, 1995 https://doi.org/10.1162/neco.1995.7.4.639
  13. L. Breirnan, Using convex pseudo-data to increase prediction accuracy, Technical report, University of Calinfornia, 1999
  14. Roberto Battiti, Anna Maria Colla, Democracy in Neural Nets: Voting Schemes for Classification, Neural Networks, 7(4), pp. 691-707, 1994 https://doi.org/10.1016/0893-6080(94)90046-9
  15. Yong Liu, Xin Yao, Simultaneous Training of Negatively Correlated Neural Networks in an Ensemble, IEEE Transactions on Systems, Man, and Cybernetics, Part B:Cybernetics, 29(6), (1999) https://doi.org/10.1109/3477.809027
  16. D. V. Sridhar, E. B. Bartlett, R C. Seagrave, An Information Theoretic Approach for Combining Neural Network Process Models, Neural Networks, 12, pp. 915-926, 1999 https://doi.org/10.1016/S0893-6080(99)00030-1
  17. Nikhil R. Pal, Vijay Kumar Eluri, Two Efficient Connectionist Schemes for Structure Preserving Dimensionality Reduction, IEEE Transactions on Neural Networks, 9(6), pp. 1142-1154, 1998 https://doi.org/10.1109/72.728358
  18. Anil K Jain, Robert P. W. Duin, Jianchang Mao, Statistical Pattern Recognition: A Review, IEEE Trans. Pattern Analysis and Machine Intelligence, 22(1), pp, 4-37, 2000 https://doi.org/10.1109/34.824819
  19. S. Lawrence, C. Lee Giles, A C. Tsoi, A D. Back, Face recognition: A convolutional neural network approach, IEEE trans. Neural Net· works, 8, pp. 98-113, 1997 https://doi.org/10.1109/72.554195
  20. M. Bichsel, A. P. Pentland, Human face recognition and the face image set's topology, CVGIP: Image Understanding, 59, pp. 254-261, 1994 https://doi.org/10.1006/ciun.1994.1017
  21. Laurenz Wiskott, Kean-Marc Fellows, Norbert Kruger, and Christoph von der Malsburg, Face Recognition by Elastic Bunch Graph Matching, IEEE Trans. On PAMI, 19(7), pp. 775-779 https://doi.org/10.1109/34.598235
  22. Chengjun Liu and Harry Wechsler, Learning the Face Space-Representation and Recognition, Proc. of the 15th International Conf. on Pattern Recognition, Vol.1 pp. 249-256 https://doi.org/10.1109/ICPR.2000.905313
  23. Baback Moghaddarn, Wasiuddin Wahid, and Alex Pentland, Beyond Eigenface: Probabilistic Matching for Face Recognition, Proc. of the 2nd International Conf. on Automatic Face and Gesture Recognition, pp. 30-35, 2000 https://doi.org/10.1109/AFGR.1998.670921
  24. Srinivas Gitta and Harry Wechsler, Modular Forensic Architecture, Face Recognition from Theory to Applications, Springer, pp. 327-347, 1997
  25. S. Gutta, J. Huang, B. Takacs, and H Wechsler, Face Recognition using Ensembles of Networks, Proc. of the 13th International Conf. on Pattern Recognition, Vol. 4, 1996 https://doi.org/10.1109/ICPR.1996.547232
  26. 평은섭, 김병화, 안현식, 김도현, 무게중심을 이용한 자동얼굴인식 시스템의 구현, 전자공학회 논문지, 제 33권, B편, 제 8호, pp. 114-122, 1996
  27. 이상영, 함영국, 박래홍, 뉴로퍼지 알고리듬을 이용한 얼굴인식, 전자공학회논문지, 제 32권, B편, 제 1권, pp. 50-63, 1995
  28. Xin Yao, and Yong Liu, Making Use of Population Information in Evolutionary Artificial Neural Networks, IEEE Trans. on Systems, Man and Cybernetics-Part B: Cybernetics, Vol. 28, No.3, 1998 https://doi.org/10.1109/3477.678637
  29. David E. Moriarty and Risto Miikkulainen, Hierarchical Evolution of Neural Networks, International Conference on Evolutionary Computation, pp. 428-433,1998 https://doi.org/10.1109/ICEC.1998.699793
  30. 류연식, 오세영, 고유얼굴과 신경망 융합을 이용한 정면얼굴의 Recall 및 인식, 제 12회 영상처리 및 이해에 관한 워크？ 논문지, pp. 528-533, 2000
  31. Daniel P. Huttenlocher, Ryan H. lilien, and Clark F. Olson View-Based Recognition using an Eigenspace Approximation to the Hausdorff Measure, IEEE Trans. on PAMI, Vol 21, No.9, pp. 951-955, 1999 https://doi.org/10.1109/34.790437