一般boosting人臉偵測演算法使用矩形特徵。為了得到較好的效能，通常需要較多的訓練樣本，這會產生無法預估的特徵數量，可能需要更多的時間來偵測人臉。此外，矩形特徵是採用像素值來計算，像素值容易受到光源的影響，因此在計算特徵值前須做前處理的動作但也會增加其運算時間。本論文提出以Adaboost演算法和類神經網路法中的倒傳遞網路為基礎，搭配區域和全域特徵以串接式架構來偵測人臉。基於階層式的人臉偵測系統，具有即時偵測且對於光源低敏感度的特性。其中區域特徵我們採用MCT特徵(Modified Census Transform Feature)，是一種紋理特徵，對光源較不敏感。使用這類紋理特徵不用對每個子視窗作前處理的動作。在挑選弱分類器上採用與一般boosting演算法不同的方法，以階層式特徵架構來控制特徵數量以減少產生過多不必要的特徵。因為MCT特徵只有描述紋理資訊，亮度資訊被移除。若單以MCT特徵計算，容易造成誤判。所以在本論文中加入全域特徵，考量亮度資訊，可以排除很多誤判區域。實驗結果顯示本論文所提架構的偵測率為99%，誤判個數11個，偵測速度為27.92 FPS。

General boosting algorithms for face detection use rectangle features. To get better performance, it needs more training samples and may generate some unpredictable number of features and that is why it needs more time to detect the face. Besides, using pixel values, which are easily affected by illumination, to calculate the rectangle features, it usually needs to preprocess the data before calculating the values of features. Such approach may increase the computation time. Our proposed solution is based on Adaboost algorithm and back propagation network of neural network combining local and global features with cascade architecture to detect human faces. This thesis presents a hierarchical face detection system with real-time operation and low sensitivity to light illumination. We use Modified Census Transform Feature (MCT), which is belonged to texture features and is less sensitive to illumination, for local feature calculation. By this approach, it is not necessary to preprocess each sub-window of the image. For classification, the selection of weak classifiers is different from that of the boosting algorithms, here we use the structure of hierarchical feature to control the number of features and it will not generate too many of them. Since MCT only describes the texture information, the brightness information is removed. With only MCT, it is very easy to misjudge faces. Therefore, in this work we include the brightness information of global features to eliminate the false positive regions. The experimental results show that the proposed architecture can have detection rate of 99%, false positives of 11, and detection speed of 27.92 FPS.

目錄
中文摘要...............................................................................................I
英文摘要..............................................................................................II
目錄....................................................................................................III
圖目錄................................................................................................VI
表目錄...............................................................................................VIII
                                                                                                                     
第一章 緒論..........................................................................................1
1.1 研究背景.........................................................................................1
1.2 研究動機與主題...............................................................................1
1.3 論文架構.........................................................................................4
第二章 人臉偵測相關研究.......................................................................5
2.1 概述................................................................................................5
2.2 人臉偵測相關文獻............................................................................7
第三章 矩形特徵及串接式分類器架構.....................................................14
3.1 概述..............................................................................................14
3.2 矩形特徵........................................................................................14
3.3 積分圖...........................................................................................17
3.4 弱分類器(矩形特徵) ........................................................................22
3.5 串接式分類器架構...........................................................................24
第四章 階層式特徵架構.........................................................................27
4.1 概述..............................................................................................27
4.2 紋理特徵........................................................................................27
4.2.1 Local Binary Pattern.....................................................................28
4.2.2 Modified Census Transform...........................................................30
4.2.3 Global Binary Pattern...................................................................33
4.3 Adaboost演算法.............................................................................35
4.3.1 演算法描述..................................................................................35
4.3.2 弱分類器(MCT) ...........................................................................36
4.4 倒傳遞類神經網路演算法.................................................................39
4.5 本論文所提的架構...........................................................................46
第五章 實驗結果與分析.........................................................................53
5.1 樣本訓練........................................................................................53
5.2 效能比較與分析..............................................................................54
第六章 結論.........................................................................................58
參考文獻..............................................................................................59

圖目錄
圖1.1  人臉光源分佈統計.........................................................................3
圖2.1  基於像素階層式特徵和統計式人臉遮罩分類器..................................8
圖2.2  基於單一特徵對映(single feature map)的人臉偵測系統....................10
圖2.3  使用粗糙-精細層(coarse-to-fine)策略，提升偵測速度.......................10
圖2.4  膚色在不同色彩空間的偵測..........................................................11
圖2.5  膚色偵測流程圖..........................................................................12
圖3.1  部分直立矩形特徵示意圖.............................................................16
圖3.2  座標(x,y)的積分圖.......................................................................17
圖3.3  矩形D的積分圖計算.....................................................................18
圖3.4  45°矩形特徵的旋轉總和區域表.....................................................20
圖3.5  旋轉45°矩形特徵.........................................................................20
圖3.6  旋轉區域的計算方法....................................................................21
圖3.7  串接式人臉偵測架構....................................................................24
圖4.1  一些不同(P,R)參數的LBP.............................................................29
圖4.2  CT與MCT編碼.............................................................................32
圖4.3  不同亮度的CT圖像.......................................................................33
圖4.4  人臉及其GBP圖像.......................................................................34
圖4.5  人臉亮度資訊分佈機率圖..............................................................34
圖4.6  最佳弱分類器位置........................................................................37
圖4.7  一般常用的BPN架構....................................................................40
圖4.8  本論文使用的BPN架構.................................................................41
圖4.9  BPN演算法學習過程示意圖..........................................................44
圖4.10  轉換函數--雙彎曲函數(Sigmoid function) ..................................... 46
圖4.11  LBP和MCT特徵比較...................................................................47
圖4.12  本論文所提的人臉偵測架構.........................................................49
圖4.13  本論文所提的串接式架構............................................................50
圖4.14  圖像金字塔................................................................................51
圖4.15  重複偵測的人臉區塊...................................................................51
圖4.16  重疊區塊示意圖.........................................................................52
圖5.1  部分ROC測試人臉圖....................................................................56
圖5.2  ROC曲線比較..............................................................................57

表目錄
表3.1  五種基本(直立)的矩形特徵............................................................15
表3.2  直立和45°矩形特徵......................................................................16
表4.1  矩形特徵和紋理特徵比較..............................................................28
表4.2  紋理特徵及其加權值....................................................................37
表5.1  本論文使用的演算法與特徵...........................................................54
表5.2  效能比較表.................................................................................55
表5.3  偵測速度比較表...........................................................................55

[1] F. R. Al-Osaimi, M. Bennamoun and A. Mian, “Spatially Optimized
Data-Level Fusion of Texture and Shape for Face Recognition,”
IEEE Transactions on Image Processing, vol. 21, no. 2, pp. 859-872, 
February 2012.
[2] P. Viola and M. Jones, “Rapid Object Detection Using a Boosted
Cascade of Simple Features,” IEEE Computer Society Conference on 
Computer Vision and Pattern Recognition, vol. 1, pp. 511-518, 2001.
[3] P. Viola and M. Jones, “Robust real-time face detection,
” International Journal of Computer Vision, vol. 57, no. 2, pp. 137-154,
May 2004.
[4] M.-H. Yang, D. Kriegman, and N. Ahuja, “Detecting faces in
images: a survey,” IEEE Transactions on Pattern Analysis and Machine 
Intelligence, vol. 24, no. 1, pp. 34-58, January 2002.
[5] K. Ali, F. Fleuret, D. Hasler and P. Fua, “A Real-Time Deformable
Detector,” IEEE Transactions on Pattern Analysis and Machine 
Intelligence, vol. 34, no. 2, pp. 225-239, February 2012.
[6] J.-M. Guo, C.-C. Lin, M.-F. Wu, and C.-H. Chang and H. Lee,
“Complexity Reduced Face Detection Using Probability-Based Face
Mask Prefiltering and Pixel-Based Hierarchical-Feature Adaboosting,”
IEEE Signal Processing Letters, vol. 18, no. 8, pp. 447-450, August 2011.
[7] H. Masnadi-Shirazi and N. Vasconcelos, “Cost-sensitive boosting,”
IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 33,
no. 2, pp. 294-309, February 2011.
[8] P. Wang, C. Shen, N. Barnes and H. Zheng, “Fast and Robust Object
Detection Using Asymmetric Totally Corrective Boosting,”
IEEE Transactions on Neural Networks and Learning Systems, vol. 23,
no. 1, pp. 33-46, January 2012.
[9] C. Shen, S. Paisitkriangkrai and J. Zhang, “Efficiently Learning a 
Detection Cascade With Sparse Eigenvectors,” IEEE Transactions on 
Image Processing, vol. 20, no. 1, pp. 22-35, January 2011.
[10] M. J. Saberian and N. Vasconcelos, “Learning Optimal Embedded 
Cascades,” IEEE Transactions on Pattern Analysis and Machine 
Intelligence, accepted.
[11] Y.-N. Chen, C.-C. Han, C.-T. Wang, B.-S. Jeng and K.-C. Fan, 
“A CNN-Based Face Detector with a Simple Feature Map and a 
Coarse-to-fine Classifier,” IEEE Transactions on Pattern Analysis and
Machine Intelligence, accepted.
[12] P. Shih and C. Liu, “Face detection using discriminating feature
analysis and Support Vector Machine,” Pattern Recognition, 
vol. 39(2), pp. 260-276, 2006.
[13] W. R. Tan, C. S. Chan, P. Yogarajah and J. Condell, “A Fusion
Approach for Efficient Human Skin Detection,” IEEE Transactions on
Industrial Informatics, vol. 8, no. 1, pp. 138-147, February 2012.
[14] L. Liu, N. Sang, S. Yang, and R. Huang, “Real-time skin color
detection under rapidly changing illumination conditions,” IEEE 
Transactions on Consumer Electronics, vol.57, no.3, pp. 1295-1302, 2011.
[15] R. Lienhart and J. Maydt, “An Extended Set of Haar-Like Features
for Rapid Object Detection,” International Conference on Image
Processing, vol. 1, pp. 900-903, 2002.
[16] 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, July 2002.
[17] B. Froba and A. Ernst, “Face Detection with the Modified Census 
Transform,” Sixth IEEE International Conference on Automatic Face 
and Gesture Recognition, pp. 91-96, May 2004.
[18] T. Ojala, K. Valkealahti, E. Oja and M. Pietikainen, “Texture
Discrimination with Multidimensional Distributions of Signed Gray 
Level Differences,” Pattern Recognition, vol. 34, no.3, pp. 727-739, 
2001.
[19] R. Zabih and J. Woodfill, “A non-parametric approach to visual
correspondence,” IEEE Transactions on Pattern Analysis and Machine 
Intelligence, 1996.
[20] Y. Freund and R. E. Shapire, “A short introduction to boosting,”
Journal of Japanese Society for Artificial Intelligence, vol. 14, no. 5, 
pp. 771-780, September 1999.
[21] R. P. Lippman, “An introduction to computing neural networks,”
IEEE ASSP Magazine, vol. 3-4, pp. 4-22, 1987.
[22] H. Jin, Q. Liu, H. Lu and X. Tong, “Face detection using improved
LBP under Bayesian framework,” Third International Conference on Image 
and Graphics, pp. 306-309, 2004.
[23] D. Huang, C. Shan, M. Ardabilian, Y. Wang and L. Chen, “Local
Binary Patterns and Its Application to Facial Image Analysis:A Survey,”
IEEE Transactions on Systems, Man, and Cybernetics—part c: Applications 
and Reviews, vol. 41, no. 6, pp. 765-781, November 2011.
[24] OpenCV (Open Source Computer Vision) is a library of programming 
functions for real time computer vision[Online]. Available: http://opencv. 
willowgarage.com/wiki/Welcome
[25] CBCL Face Database #1[Online]. Available: http://cbcl.mit.edu/cbcl/
software-datasets/FaceData2.html
[26] BioID Face Database[Online]. Available: http://www.bioid.com/ 
downloads/software/bioid-face-database.html
[27] CALTECH Face Database [Online]. Available: http://www.vision.
caltech.edu/html-files/archive.html
[28] AT&T Face Database[Online]. Available: http://www.cl.cam.ac.uk/
research/dtg/attarchive/facedatabase.html
[29] gray FERET Face Database[Online]. Available: http://www.itl.nist.gov 
/iad/humanid/feret/feret_master.html
[30] MUCT Face database(Images from camera a)[Online]. Available: http://www.milbo.org/muct/