本論文提出一種具光影變化抗性的人臉特徵擷取技術與一種改良式辨識決策技術並將兩者結合，實現一套即時人臉性別辨識系統。近年來人臉偵測技術日趨成熟，因此針對人臉影像的各種辨識研究也越受矚目，人臉性別辨識是其中倍受重視的一項研究，可應用於行銷廣告、人口統計、監控領域中。
在人臉特徵擷取方面，由於特徵描述的穩定度與強建性，將會最直接呈現在系統辨識率上，考慮到人臉上的紋理與對比結構精細度大小不一，因此本論文提出基於參數導向直方圖等化之矩形特徵技術，可對人臉上不同區域大小的特徵進行強化與描述，並同時具備光影變化抗性，此技術運算複雜度低可供系統快速取得強建而穩定的人臉性別特徵進行辨識。
在性別辨識方面，傳統的AdaBoost分類器在使用上雖然有著高處理效率，但是弱分類器之間在進行決策時並無互相分享補償的機制，造成某些弱分類器決策錯誤發生時無法彌補，從而影響至最終強分類器的投票率，導致辨識率下降。針對此一問題本論文提出基於點擴散之AdaBoost決策權重分享技術，對AdaBoost分類器在決策過程中進行處理，於弱分類器分佈空間中對相似的弱分類器彼此間進行分享的動作，以彌補傳統AdaBoost的不足。
實驗結果方面，本論文使用FERET與LFW此二公開人臉資料庫各別進行測試，並與前人的方法進行比較，從結果可看出本論文提出的人臉性別辨識架構其不論系統辨識率與處理效能皆是相當突出的。最後為了將所學回饋於現實生活之中，本論文將人臉性別辨識架構結合人臉偵測技術(AdaBoost)，以筆記型電腦、PTZ攝影機、USB影像擷取卡模擬出一套介面化即時人臉性別辨識系統。

This thesis proposed two technologies to integrate a real-time face gender recognition system. One of the proposed techniques is face feature extraction which has illumination invariance, and the other one presents an improved strategy of recognition technique. Recently, the face detection related technologies has been improving, thus the research for face recognition with various fields also became popular. Face gender recognition is one of the most popular research fields in pattern recognition, which can be applied in advertisting, marketing, demographics and momitoring system.
In terms of face feature extraction, the feature’s stability and robustness are the key issues on system recognition rate. Considering the degrees of texture and regional contrast on human face are different, this thesis proposed the Parametric-Oriented Histogram Equalization based Haar-like Feature (POHEH) to enhance and describe the facial features with various regional scales. The proposed method also gears with illumination invariance property. As the computational complexity of the POHEH is low, it can efficiently provide a robust and stable facial gender feature to the face gender recognition system.
In terms of gender recognition, in fact, the traditional Adaboost cassifier has high efficiency in processing, though there is no sharing compensation mechanism between weak classifiers in predicting, leading to remedied unablely conditions when a certain weak classifiers induce error decision. To solve this problem, this thesis proposed the Dot-Diffusion-based AdaBoost (DDAdaBoost) which can propagate the decision error of a weak classifier to its neighboring classifiers in ficitious space. As expected, the DDAdaBoost shows a promising result compared to the typical of Adaboost.
In the experimental results, the two standard databases, FERET and LFW, are considered, and improvements are demonstrated in terms of the recognition accuracy rate and the efficiency of the overall system. Both of the recognition rate and the processing efficiency of the proposed system are superior to the former researches. Finally, the equipments, notebook, PTZ camera and USB video capture card, are employed to construct a real-time face gender recognition system.

[1] D. Gabor, “Theory of Communications,” J. Institution of Electrical Engineers, vol. 93, pp. 429-457, 1946.
[2] J. G. Daugman, “Uncertainty Relation for Resolution In Space, Spatial- Frequency, and Orientation Optimized by Two-Dimensional Visual Cortical Filters,” J. Optical Society of America A-Optics Image Science and Vision, vol. 2, pp.1160-1169, 1985.
[3] Q. Y. Zhao, B. C. Pan, J. J. Pan and Y. Y. Tang, "Facial expression recognition based on fusion of Gabor and LBP features ," International Conference on Wavelet Analysis and Pattern Recognition, vol. 1, pp. 362-367, Aug. 2008.
[4] S. M. Lajevardi, M. Lech, "Facial Expression Recognition Using Neural Networks and Log-Gabor Filters," Digital Image Computing: Techniques and Applications, pp. 77-83, Dec. 2008.
[5] J. F. Ye, Y. Z. Zhan and S. L. Song, "Facial expression features extraction based on Gabor wavelet transformation," IEEE International Conference on System, Man and Cybernetics, vol. 3, pp. 2215-2219, Oct. 2004.
[6] W.F. Liu, Z.F. Wang, "Facial Expression Recognition Based on Fusion of Multiple Gabor Features," 18th International Conference on Pattern Recognition, vol. 3, pp.536-539, Sep. 2006.
[7] T. Ojala, M Pietikinen and D. Harwood, "A comparative study of texture measures with classification with local binary patterns," Pattern Recognition, vol. 29, no. 1, 1996.
[8] T. Ojala, M. Pietikainen and T. Maenpaa, “Multiresolution gray-scale and rotation invariant texture classification with local binary patterns,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, no. 7, pp. 971-987, Jul.2002.
[9] G. Shakhnarovich, P. A. Viola and B. Moghaddam, "A unified learning framework for real time face detection and classification," Fifth IEEE International Conference on Automatic Face and Gesture Recognition, pp.14-21, May 2002.
[10] P. Viola and M. J. Jones, “Robust real-time object detection,” Second International Workshop on Statistical and Computational Theories of Vision, July.2001.
[11] M. Isard and A. Blake, "Condensation – conditional density propagation for visual tracking," International Journal of Computer Vision, Vol. 29, No.1, pp. 5–28, 1998.
[12] B. Moghaddam and M. H. Yang, "Gender classification with support vector machines," Fourth IEEE International Conference on Automatic Face and Gesture Recognition, pp. 306–311, Mar. 2000.
[13] B. Moghaddam and M. H. Yang, “Learning gender with support faces,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 24, no. 5, pp. 707-711, May 2002.
[14] B. Moghaddam and A. Pentland, "Probabilistic visual learning for object representation," IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), vol. 19, no. 7, pp. 696-710, July 1997.
[15] H. Lu and H. Lin, “Gender recognition using AdaBoosted feature,” Third International Conference on Natural Computation, vol. 2, pp. 646-650, Aug. 2007.
[16] M. J. Lyons, J. Budynek, A. Plante and S. Akamatsu, "Classifying facial attributes using a 2-D Gabor wavelet representation and discriminant analysis," Fourth IEEE International Conference on Automatic Face and Gesture Recognition, pp. 202-207, Mar. 2000.
[17] S. U. Jung, D. H. Kim, K. H. An and M. J. Chung, "Efficient rectangle feature extraction for real-time facial expression recognition based on AdaBoost," IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 1941-1946, Aug. 2005.
[18] E. Makinen and R. Raisamo, "An experimental comparison of gender classification methods," Pattern Recognition Letters, vol. 29, no. 10, pp. 1544-1556, July 2008.
[19] H. C. Lian and B. L. Lu, "Multi-view gender classification using local binary patterns and support vector machines," Third International Symposium on Neural Networks (ISNN), vol. 3972, pp. 202–209, May 2006.
[20] B. Wu, H. Ai and C. Huang, "LUT-based AdaBoost for gender classification," International Conference on Audio- and Video-Based Biometric Person Authentication (AVBPA), vol. 2688, pp. 104–110, June 2003.
[21] B. Wu, H. Ai and C. Huang, "Real-time gender classification," Third International Symposium on Multispectral Image Processing and Pattern Recognition, vol. 5286, pp.498-503, Oct. 2003.
[22] C. Shan, “Learning local binary patterns for gender classification on real-world face images,” Pattern Recognition Letters, vol. 33, no. 4, pp. 431 - 437, March 2012.
[23] C. P. Papageorgiou, M. Oren and T. Poggio, “A general framework for object detection,” Sixth International Conference on Computer Vision, pp. 555-562, Jan. 1998.
[24] R. Lienhart, A. Kuranov and V. Pisarevsky, “Empirical analysis of detection cascades of boosted classifiers for rapid object detection,” Pattern Recognition, vol. 2781, pp. 297-304, Sep. 2003.
[25] D. H. Kim , S. U. Jung and M. J. Chung, “Extension of cascaded simple feature based face detection to facial expression recognition,” Pattern Recognition Letters, vol. 29, no. 11, pp. 1621-1631, Aug. 2008.
[26] R. C. Gonzalez and R. E. Woods, Digital Image Processing, 2nd ed., Reading, MA: Addison-Wesley, 1992.
[27] Y. F. Liu, J. M. Guo and B. S. Lai, “High Efficient Contrast Enhancement Using Parametric Approximation,” International Conference on Acousties, Speech and Signal Processing, Vencouver, Canade, May 26-31, 2013.
[28] M. Abramowitz and I. A. Stegun, Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables, NY: Dover, 1964.
[29] D. E. Knuth, “Digital halftones by dot diffusion,” ACM Trans. Graph., vol. 6, no. 4, Oct. 1987.
[30] Y. Freund and R. E. Schapire, “A decision-theoretic generalization of on-line learning and an application to boosting,” Journal of Computer and System Sciences, vol. 55, no. 1, pp.119-139, Aug.1997.
[31] National Institute of Standards and Technology: The Color FERET Database. http://www.nist.gov/itl/iad/ig/colorferet.cfm
[32] Labeled Faces in the Wild (LFW): http://vis-www.cs.umass.edu/lfw/
[33] X. Tan and B. Triggs, “Enhanced Local Texture Feature Sets for Face Recognition Under Difficult Lighting Conditions,” IEEE Trans. Image Processing, vol. 19, no. 6, pp. 1635-1650, June 2010.
[34] C. Shan, “Smile Detection by Boosting Pixel Differences,” IEEE Trans. Image Processing, vol. 21, no. 1, pp. 431-436, Jan. 2012.
[35] SONY EVI-D70技術手冊： http://pro.sony.com/bbsc/ssr/product-EVID70P/
[36] Open Source Computer Vision (OpenCV): http://opencv.org/
[37] S. Balija and H. A. Rowley, “Boosting sex identification performance,” International Journal of Computer Vision, vol. 71, no. 1, pp. 111-119, 2007
[38] X. M. Leng and Y. D.Wang, “Improving generalization for gender classification” IEEE International Conference on Image Processing, pp.1656-1659, Oct. 2008
[39] E. Makinen and R. Raisamo, “Evaluation of gender classification methods with automatically detected and aligned faces, ”IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 30, no. 3, pp. 541–547, March 2008.
[40] M. Toews and T. Arbel, “Detection, Localization, and sex classification of faces from arbitrary viewpoints and under occlusion, ”IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 31, no. 9, pp. 1567-1581, 2009
[41] J. Bekios-Calfa, J. M. Buenaposada and L. Baumela, “Revisiting linear discriminant techniques in gender recognition,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 33, no.4, pp. 858-864, 2011