人臉偵測演算法不斷的被提出並實現為即時臉部偵測系統，且廣泛地商業化其應用範圍，但並非所有系統都能穩健的應用於照度變化下的環境。首先本論文以標準人臉資料庫評估一現有即時臉部偵測系統，其研究目的在於找出該系統效能可提升之處，經評估後，該系統在特定照度下無法偵測，且照度變化的環境下有太高的錯誤拒絕率。因此本論文提出以多形態的臉部樣板解決廣泛的照度變化，並與預設樣板結合做為臉部偵測器，使用標準人臉資料庫訓練分類器，並以靠近邊限上的樣本來調整分類器的效能，採用蒙地卡羅模擬法則評估隨機選取樣本對於效能的影響。

Face detection is nowadays applied in many devices, and generally considered as a mature technology. But some performance weakness may be discovered under a strict examination. Given a real-time face detector, this thesis aims at identifying its weakness at detecting faces under various illuminations, and proposing a solution to compensate for the weakness. This work first exploits a benchmark database such as PIE and FRGC to identify the illumination conditions that the face detector yields a high miss rate. When these illumination conditions are identified, a multi-template search scheme is proposed so that facial candidates with different illuminations can be considered. To meet the requirements for different needs, this work exploits a combination of multiple classifiers and the training samples close to the decision boundaries, making the performance easy to tune for a lower FAR (False Accept Rate) or FRR (False Reject Rate). Experiments show that the multi-template search scheme can improve the detection rate from 72 to 90 percentage, and the multiple classifiers can improve the FAR from 0.028 to 0.012, at FRR 0.05.

[1]J.P. Lewis Fast Template Matching Vision Interface 95, Canadian Image Processing and Pattern Recognition Society, pp. 120-123, 1995
[2]H. A. Rowley, S. Baluja & T. Kanade Neural Network-Based Face Detection IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 20, no. 1, pp. 23-38, 1998
[3]Kah-Kay Sung, Tomaso Poggio Example-Based Learning for View-Based Human Face Detection IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 20, no. 1, 1998
[4]P.J. Phillips, P.J. Rauss & S.Z. Der FERET (Face Recognition Technology) Recognition Algorithm Development and Test Results Technical Report ARL-TR-995, Army Research Lab, 1996
[5]M. H. Yang, D. Kriegman & N. Ahuja Detecting Faces in Images: A survey IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 24, no. 1, pp. 34-58, 2002
[6]Clyde Shavers, Robert Li & Gary Lebby An SVM-based Approach to Face Detection System Theory, 2006 Proceeding of the Thirty-Eighth Southeastern Symposium on Publication 5-7, 2006
[7]M. Pontil, B. Heisele, T. S. & T. Poggio Component-based face detection IEEE Int. Conf. Computer Vision and Pattern Recognition, pp 657-662, 2001
[8]M. Pontil, B. Heisele, T. S. & T. Poggio A Component based framework for face detection and identification International Journal of Computer Vision, 2006
[9]Paul Viola & Michael Jones Rapid object detection using a boosted cascade of simple features Proc. IEEE Conf. Computer Vision and Pattern Recognition, vol.1, pp. 511-518, 2001
[10]R. Lienhart & J. Maydt An Extended Set of Haar-like Features for Rapid Object Detection IEEE International Conference on Image Processing, 2002
[11]Schneiderman, H. & T. Kanade A Statistical Method for 3D Object Detection Applied to Faces and Cars Proc. IEEE Conf. Computer Vision and Pattern Recognition, 2000
[12]T. Sim, S. Baker & M. Bsat The CMU Pose, Illumination, and Expression (PIE) Database Proc. IEEE International Conference on Automatic Face and Gesture Recognition, 2002
[13]Phillips, P. J. & et. al. Overview of the Face Recognition Grand Challenge Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2005
[14]Reinhard Klette & Azriel Rosenfeld Digital geometry, geometric methods for digital picture analysis California, USA: Morgan Kaufmann, 2004, pp. 455-47
[15]Richard O. Duda, Peter E. Hart, D. G. S. Pattern Classification Wiley-Interscience, 2000
[16]林隆煥 著 視窗程式設計函式庫-Win 32 API（上） 金禾資訊, 2004
[17]林隆煥 著 視窗程式設計函式庫-Win 32 API（下） 金禾資訊, 2004
[18]Kasturi, R. & et. al. Framework for performance evaluation of face, text, and vehicle detection and tracking in video: Data, metrics, and protocol IEEE Trans. Pattern Analysis and Machine Intelligence, pp. 1-17, 2003