(同濟大學電子與信息工程學院  上海  201804)

摘要  利用人臉作為特征的生物識別系統是近年來模式識別和圖像處理領域的研究熱點之一。介紹了一種改進的人臉識別算法。算法以主成分分析（PCA）算法作為主體，以AdaBoost算法作為輔助，把以投影后人臉特征空間中的歐式距離作為識別的主要評判依據。與傳統人臉識別系統相比，新算法可以避免系統在識別前進行人臉檢測的巨大運算量，并有效區分人臉和非人臉圖像，提高運算效率和識別精度。仿真結果表明，這種改進的算法硬件資源占用少，運算時間短，更適合在嵌入式平臺上實現。

關鍵詞 人臉識別；人臉檢測；主成分分析；MATLAB仿真

中圖法分類號 TP391

An Improved Face Recognition Algorithm Based on PCA and AdaBoost

ZHANG Xu-dong,  XU He-gen
(School of Electronics and Information, Tongji University, Shanghai, 201804)

Abstract  In recent years, the biometric identification system using human face as a biometric characteristic is one of the highlights in agro-scientific research in the field of pattern recognition and image processing. An improved face recognition algorithm is proposed in the present paper. In this algorithm, the principle component analysis (PCA) algorithm is the main body, the AdaBoost algorithm is the auxiliary, the Euclidean distance in projected face feature space is the key indicator of recognition. Compared with the traditional face recognition system, the algorithm can avoid the huge computational complexity in the face detection phase before recognition phase. It also can distinguish between face and non-face images effectively and improve the operational efficiency and accuracy. The simulation results show that this improved algorithm has less hardware resource usage, shorter operation time and it is more suitable to implement on the embedded platform compared with the traditional face recognition system.

Keywords  face recognition; face detection; principle component analysis; simulation in MATLAB

參考文獻 (references)
[1] Turk M, Pentland A. Eigenfaces for recognition [J]. Cognitive Neurosci, 1991, 3(1): 71-79.
[2] Freund Y, Schapire R E. A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting [J]. Journal of Computer and System Sciences, 1997, 55(1): 119-139.
[3] Samal A, Iyengar P A. Automatic Recognition and Analysis for Human Faces and Facial Expressions: A Survey [J]. Pattern Recognition, 1992, 25(1): 65-77.
[4] Chellappa R, et al. Humnan and machine recognition of face: a survey [J]. Proc. IEEE, 1995, 83(5): 705-740.
[5] Jang Kyounghoon, Cho Hosang, Kim Changwan, Kang Bongsoon. A Face-detection Postprocessing Scheme using a Geometric Analysis for Multimedia Applications [J]. Journal of Semiconductor Technology and Science, 2013, 13(1):35-42.
[6] Lee R. Face Recognition Elasitc Relation Encoding and Structural Matching [J]. Proceedings of the IEEE international Conference on Systems, Men and Cybernetics, 1999, 172-177.
[7] Lawrence S, Giles C L, tsoi A C. Face recognition: A convolutional neural network approach [J]. IEEE Transactions on Neural Networks, 1997, 8(1): 98-113.
[8] 俞燕, 李正明. 基于特征的彈性圖匹配人臉識別算法改進 [J]. 計算機工程, 2011, 37(5): 216-218.
[9] 王金輝. 人臉識別算法研究 [D]. 西安電子科技大學, 2010.
[10] Viola P, Jones M J. Robust Real-Time Face Detection [J]. International Journal of Computer Vision, 2004, 57(2): 137-154.
[11] Sirovich L, Kirby M. Low-dimensional procedure for characterization of human faces [J]. Optical Soc Am, 1987, 4(3): 519-524.
[12] Kirby M, Sirovich L. Application of the karhunen-loeve procedure for the characterization of human faces [J]. IEEE PAMI, 1990, 12(1): 103—108.
[13] 鄧楠. 基于主成分分析的人臉識別研究 [D]. 西北大學, 2006.

作者簡介
張旭東，男，1987年生，碩士研究生二年級，現就讀于上海同濟大學電子與信息工程學院，檢測技術與自動化裝置專業，主要研究方向為MATLAB圖像處理仿真和基于FPGA的圖像處理。