人臉偵測與辨識一直廣泛的應用在不同領域，本論文針對Smart Phone，以跨平台Qt GUI開發環境整合人臉偵測與辨識模組，並以加裝攝影機的嵌入式Linux平台模擬Smart Phone，可即時完成偵測與辨識使用者，確保機密不外洩。

本系統主要包含二個子模組，分別為人臉偵測模組和人臉訓練與辨識模組，人臉訓練與辨識的部份，採用進階相關性過濾器(Advanced Correlation Filters)來實做，並因應嵌入式平台速度與記憶體的限制，選擇適合的過濾器，將人臉資料透過快速傅立葉轉換轉至頻率域，並透過進階相關性過濾器以一人一檔案的方式製作成XML格式的人臉資料檔，以支援多人共用的Smart Phone裝置。

系統經過實際測試，可正確的完成偵測與辨識使用者，使用ESL人臉資料庫，辨識率達92.57%，於嵌入式系統運作的效率在XScale-PXA270 520 MHz嵌入式系統環境下，單一人臉偵測速度可達200ms，單一人臉辨識速度可達400ms。

The face detection and recognition are always applied to many different fields. This thesis focuses on the Smart Phones. We combine the human face detection and recognition modules by using Qt GUI Integration Development Environment (IDE) and utilize embedded system platform which has camera emulating Smart Phone. It can work in real-time and prevent the secret data be stolen via impostors.
The system is composed of two modules: human face detection and human face database training and recognition. About the modules of human face database training and recognition, we implement those by using the Advanced Correlation Filters and choose the suitable one for the embedded system constraints in speed and memory. We convert the human face data from spatial domain to frequency domain by Fast Fourier Transform (FFT) in order to build human face database via Advanced Correlation Filter. For multi-user Smart Phone, we store human database in XML format file for each user.
The experiment result shows that our system can detect human face and recognize it in real-time. The recognition rate is 92.57% by using ESL human face database. The performance on embedded system (XScale-PXA270 520 MHz) is 200ms for detection and 400ms for recognition per face.

[1]B.V.K. Vijaya Kumar, M. Savvides, K. Venkataramani, Xie Chunyan. “Spatial frequency domain image processing for biometric recognition Image Processing,” IEEE ICIP, vol 1, pp.22-25, Sept. 2002.
[2]M.H. Yang, D. Kriegman, and N. Ahuja. “Detecting faces in images: A survey,” IEEE Trans. on Pattern Anal. Mach. Intell., 24(1), 34-58, 2002.
[3]Paul Viola, Michael J. Jones. “Robust Real-time Object Detection,” Cambridge Research Laboratories, 2001/01.
[4]Y. Freund and R.E. Schapire. “A decision-theoretic generalization of no-line learning and an application to boosting,” Journal of Computer and System Sciences, vol.55, pp. 119-139, 1997.
[5]M.A. Turk and A.P. Pentland, Face recognition using eigenfaces, Computer Vision and Pattern Recognition, 1991.
[6]Peter N. Belhumenr, Joao P. Hespanha and David J. Kriegman. “Eigenfaces vs. Fisherfaces:Recognition Using Class Specific Linear Projection,”IEEE Transactions on Pattern Analysis and Machine Intellgence, vol. 19., No. 7, July 1997.
[7]Wikipedia, “Discrete Fourier transform,” http://en.wikipedia.org/wiki/Discrete_Fourier_transform
[8]Marios Savvides. Introduction to Biometric Recognition Technologies and Applications, Carnegie Mellon CyLab & ECE.
[9]探矽工作室， ARM原理與實作-以Soc為例，宏友圖書，中華民國92年。
[10]Andres Calderon and Nelson Castillo. “Why ARM's EABI Matters,”
http://www.linuxdevices.com/articles/AT5920399313.html, 14, Mar 2007.
[11]Wikipedia, “Qt(toolkit),”http://en.wikipedia.org/wiki/Qt_(toolkit)
[12]K. Jensen and D. Anastassiou, “Subpixel edge localization and the interpolation of still images,” IEEE Trans. On Image Processing. vol. 4, pp. 285-295, Mar 1995.
[13]M. Sonka, V. Hlavac, and R. Boyle, Image Processing, Analysis and Machine Vision, PWS Publishing, pp. 67-68, 1999.
[14]劉鴻明，「Splines應用於影像內插法之研究」，中原大學資訊工程學系碩士論文，2005。
[15]A. Mahalanobis, B.V.K. Vijaya Kumar, and D. Casasent. “Minimum average correlation energy filters,” Applied Optics, vol. 26, pp. 3630-3633, 1987.
[16]B.V.K. Vijaya Kumar, A. Mahalanobis, S.R.F. Sims, J. F. Epperson. “Unconstrained correlation filters,” Applied Optics, vol. 33, pp. 3751-3759, 1994.
[17]M. Savvides, B.V.K. Vijaya Kumar and P. Khosla. “Face verification using correlation filters,” Proceeding of Third IEEE Automatic Identification Advanced Technologies, pp. 56-61, 2002.
[18]M. Savvides, K. Venkataramani and B.V.K. Vijaya Kumar. “Incremental Updating of Advanced Correlation Filters for Biometric Authentication System,” Proceeding of ICME, vol. 3, pp. 229-232, 2003.
[19]P. Hennings, B.V.K. Vijaya Kumar. “Palmprint Recognition Using Correlation Filter Classifiers,” Conference Record of The 38th Asilomar Conference on Signal, Systems and Computers, vol. 1, pp. 567-571, 2004.
[20]Wikipedia, “Standard deviation,” http://en.wikipedia.org/wiki/Standard_deviation
[21]張家豪，「以AAM與PCA為基礎之眼鏡特徵弱化方法於人臉辨識之改進」，中央大學資訊工程學系碩士論文，2008。
[22]楊煒達，「簡易方法之少量人臉辨識系統」，中央大學資訊工程學系碩士論文，2007。
[23]井上誠喜、八木伸行、林 正樹、中須英府輔、三谷公二、奧井誠人，C語言數位影像處理，全華圖書，中華民國95年。
[24]R. C. Gonzalez、R. E. Woods、S. L. Eddins，數位影像處理-運用MATLAB，東華書局，中華民國97年。
[25]Open Source Computer Vision Library,
http://www.intel.com/technology/computing/opencv/
[26]Christopher Hallinan，Embedded Linux 開發實務徹底研究，旗標出版社，中華民國96年。
[27]S.Rodriguez、Fischer、Smolski，The Linux Kernel Primer，上奇出版社，中華民國95年。
[28]劉翁昌，「複雜環境下之即時人臉偵測與辨識系統」，台灣科技大學電子工程學系碩士論文，2009。