本篇論文提出一個新方法來辨識人臉，由於在人臉辨識研究當中，比較難以克服的問題是人臉影像變化太急劇，這樣的問題在一般的辨識方法上都不會有太高的辨識率，因此本篇論文即是在克服這樣的問題。針對這樣的問題，在資料的輸入上就必須改變方法，一般都是利用整張人臉來做分析，但是光線及角度變化在整張人臉上的影響甚巨，所以在此篇論文中,資料的輸入必須加以考慮，因此在本篇論文中利用人臉的重要特徵區域來代替整張人臉。至於分析的方法則是利用獨立成分分析法，這個方法的好處是可以有效克服人臉在光線及角度變化下的辨識，除了主要的獨立成分分析方法之外，另外也加入主成分分析的應用當成獨立成分分析的前處理，而最後的相似度計算也提出了新的計算方式，因而成就了這樣的一個辨識方法。至於實驗中的訓練影像與測試影像資料庫，更利用了一般國際上研究所公認的FERET 人臉資料庫與實驗室的人臉資料，在這樣的方法之下，實驗的結果也如預期得到不錯的辨識率。

　　In this paper I bring up a new method to recognize human face. What is more difficult to overcome in human facial recognition is the sharp change of facial image. The ordinary recognition method cannot achieve a high recognition rate. This paper is to discuss how to overcome this problem. For this problem, the data input needs some changes. The ordinary method is to analyze a whole face, but the changes of light and angle greatly effect the whole face. Therefore, the data input must be added into consideration. In this paper I’ll use facial features instead of the whole face. Then, I’ll use Independent Component Analysis method for analysis. The advantage of the recognition method is to overcome the changes of the light and the angle on the human face. In addition to this method, I also apply Principle Component Analysis method as a pretreatment for Independent Component Analysis. At last I bring up a new calculation formula for calculating the facial similarities. Thus, I construct such a recognition method. Furthermore, I use internationally accredited (FERET) database and the human facial data collected from the experimental room as imagery rehearsal and test image database. Thus, the experimental result has a high recognition rate as expected.

[1] M.S. Bartlett and T.J. Sejnowski, “Independent component
representations for face recognition,” SPIE Conf. On Human Vision and
Electronic Imaging Ⅲ , vol. 3299, pp. 528-539, San Jose, Jan. 1998.
Recently appeared in M.S. Bartlett, J.R. Movellan and T.J. Sejnowski, “Face
recognition by ICA,” IEEE Trans. on Neural Networks, vol. 13, no. 6, Nov.
2002.

[2] P. Belhumeur, P. Hespanha, D. Kriegman, “Eigenfaces vs fisherfaces:
recognition using class specific linear projection,” IEEE Transactions on
Pattern Analysis and Machine Intelligence,Volume: 19 Issue: 7 ,July 1997
Page(s): 711-720.

[3] A. Bell and T. Sejnowski, “An information maximization approach to
blind separation and blind deconvolution,” Neural Computation 7, Nov 1995,
Page(s): 1129-1159.

[4] D.J. Beymer, “Face recognition under varying pose,” In Proceedings
IEEE Conf. on Computer Vision and Pattern Recognition, Seattle, WA, 1994,
Page(s): 756-761

[5] S.M. Bileschi, B. Heisele, “Advances in component-based face
detection, ” in: Proc. of Pattern Recognition with Support Vector Machines,
First International Workshop, SVM 2002, Niagara Falls, 2002, Page(s):
135-143.

[6] R. Brunelli, T. Poggio, “Face recognition: features versus templates,”
IEEE Trans. Pattern Anal. Mach. Intell. 15, no. 10, 1993, Page(s):1042-1052.

[7] L. Chen, H. Liao, M. Ko, J. Lin, and G. Yu, “A new LDA-based face
recognition system which can solve the small sample size problem,” Pattern
Recognition, Vol. 33, no. 10, 2000, Page(s): 1713-1726.

[8] P. Comon, “Independent component analysis – a new concept?,”
Signal Processing 36, 1994, Page(s):287-314.

[9] K. Etemad, and R. Chellappa, “Discriminant Analysis for Recognition
of Human Face Images,” J. Opt. Soc. Amer. A, Vol. 14, 1997, Page(s):
1724-1733.

[10] B. Heisele, P. Ho, and T. Poggio, “Face Recognition with Support
Vector Machines: Global versus Component-based Approach,” In Proc. IEEE
International Conference on Computer Vision, Vol. 2, 2001, Page(s): 688-694

[11] B. Heisele, T. Poggio, M. Pontil, “Face detection in still gray images,”
AI Memo 1687, Center for Biological and Computational Learning, MIT,
Cambridge, MA, 2000.

[12] J. Huang, V. Blanz, B. Heisele, “Face recognition using
component-based svm classification and morphable models,” in: Proc. of
Pattern Recognition with Support Vector Machines, First International
Workshop, SVM 2002, Niagara Falls, 2002, Page(s): 334-341.

[13] H. Kim, D. Kim, S. Bang, “An efficient model order selection for
PCA mixture model.” Pattern Recognition Letters 24(9-10): (2003), Page(s):
1385-1393.

[14] H. Kim, D. Kim, S. Bang, “Face Retrieval Using 1st- and 2nd-order
PCA Mixture Model,” International Conference on Image Processing,
Rochester, New York, Sept. 2002. Page(s): 391-400.

[15] S. Lawrence, P. Yianilos, I. Cox, “Face recognition using
mixture-distance and raw images,” Proceedings of the International
Conference on IEEE System, Man, and Cybernetics, 1997, Page(s):
2016-2021.

[16] B. Moghaddam, and A. Pentland, “Probabilistic visual learning for
object representation,” IEEE Trans. On PAMI, vol. 19, no. 7, July, 1997,
Page(s): 696-710.

[17] C. Nakajima, M. Pontil, T. Poggio, “People recognition and pose
estimation in image sequences,” Proc. IEEE-INNS-ENNS International Joint
Conf. On Neural Networks, 2000, Vol. 4, Page(s): 4189-4195.

[18] H. O’Toole, Abdi, K. Deffenbacher, and D. Velantin,
“Low-dimensional representation of faces in higher dimensions of the face
space.,” Journal of the Optical Society of America A 10(3), Page(s): 405-411,
1993.

[19] M. Turk and A. Pentland, “Eigenfaces for recognition,” J. Cog.
Neurosci. 3(1), Page(s): 71-86, 1991.