臺灣博碩士論文加值系統

摘 要
近年來，在各種影像處理之研究有關人臉偵測（Face Detection）的探討非常多，尤其在即時人臉追蹤與身份辨識等相關之研究，人臉偵測往往是必要的前置作業。一般人臉偵測可供參考的資訊很多，例如色彩資訊（Color Information）及五官特徵等。然而在一張影像中偵測人臉位置，經常受到臉部的角度、肢體的位置和五官特徵不明顯等因素的影響，導致偵測結果有所偏差。本文提出一個人臉偵測的方法，主要是先依據膚色特有的色彩資訊判斷膚色區域，接著藉由五官特徵(如眼睛、嘴巴)區域的色彩特性，取得我們所要的特徵座標，最後再根據人臉邊緣的形狀與特徵座標分布等條件，來判定影像中的人臉位置。由實驗結果顯示，本文之人臉偵測方法無論在準確度和運算時間上都有不錯的結果，相信對後續之即時人臉追蹤與身份辨識等相關研究，會有不錯的參考與幫助。

Abstract
In recent years, the relative research of face detection in image processing territory is very plenty. Especially, face detection is often a prior process for real time face tracking or recognition. A lot of information can be used for face detection, such as color information, facial features, and so on. But detecting face positions in an image is often influenced by the face direction, the limb position, and the ambiguous face features. In this thesis, we propose a method that is first according to skin’s color information to judge skin’s areas and then get the coordinates of face features through their color characteristics. Finally, the face position in the image is judged bsed on the edge of face area and the position of feature’s coordinates. The experimental results show that the use of our method in face detection can achieve a not bad job in both the accuracy and the operating time. Thus, it can provide a good reference and is helpful for the following face tracking or recognition.

目 錄
中文摘要 i
英文摘要 ii
謝誌 iii
目錄 iv
圖目錄 vii
表目錄 x
第一章 緒論 1
1.1 研究動機與目的 1
1.2 文獻探討 1
1.3 研究方法及步驟 2
1.4 論文架構 3
第二章 膚色與臉部特徵之偵測 4
2.1 膚色分割 4
2.1.1 色彩空間轉換 5
2.1.2 膚色分割門檻 9
2.2 補洞演算法 10
2.3 色彩資訊過濾 14
2.3.1 色彩資訊門檻設定 14
2.4 標記連通法 18
2.5 小結 23
第三章 臉部範圍之偵測 24
3.1 厚邊緣影像 24
3.1.1 膨脹與邊緣處理 25
3.1.2 厚邊緣偵測 28
3.2 橢圓遮罩偵測 30
3.3 臉部條件偵測 32
3.3.1 方形遮罩偵測 32
3.3.2 臉部特徵座標設定 35
3.4 小結 41
第四章 模擬結果與討論 42
4.1 實驗配備 42
4.2 模擬結果 42
4.3 討論 56
第五章 結論與展望 60
5.1 結論 60
5.2 展望 60
參考文獻 61

圖目錄
圖1.1 研究流程圖 3
圖2.1 本章流程架構圖 4
圖2.2 HSI色彩空間的模型架構 7
圖2.3 色彩空間轉換 8
圖2.4 圖2.3(a)的三分量 9
圖2.5 轉換圖2.4至YCrCb的影像顯示 9
圖2.6 改變亮度門檻的差異 10
圖2.7 補洞演算法範例 12
圖2.8 模擬影像的補洞演算 14
圖2.9 圖2.8的白色區域的原始色彩值呈現 15
圖2.10 影像的雜訊濾除結果 18
圖2.11 連通遮罩 18
圖2.12 特徵影像的特徵中心座標群 20
圖2.13 以特徵中心座標群判斷影像的人臉區域 21
圖2.14 特徵座標判斷人臉的問題影像 22
圖3.1 本章流程架構圖 24
圖3.2 單位遮罩設定圖 25
圖3.3 膨脹運算範例 26
圖3.4 侵蝕運算範例 27
圖3.5 邊界抽取範例 28
圖3.6 厚邊緣偵測 30
圖3.7 橢圓遮罩 30
圖3.8 橢圓臉型內的特徵分布 33
圖3.9 方形遮罩偵測 34
圖3.10 方形遮罩偵測的範例 35
圖3.11 方形遮罩偵測錯誤的情形 36
圖3.12 修正方形遮罩偵測的結果 37
圖3.13 (a)以方形遮罩對圖3.9(b)偵測成功的區域，(b)人臉偵測的最終分布，(c)圖(b)的原始色彩值重現 39
圖3.14 (a)將圖2.13(a)做膨脹處理，(b)圖2.13(a)的厚邊緣影像，(c)方形遮罩偵測圖(b)的人臉區域，(d)圖(c)的膚色原始色彩重現 41
圖4.1 單人模擬 44
圖4.2 雙人(前後)模擬 46
圖4.3 多人模擬 48
圖4.4 側臉模擬 51
圖4.5 斜臉模擬 53
圖4.6 黑人膚色模擬 56

表目錄
表2.1 特徵像素的RGB分量範圍 16
表4.1 其他模擬的統計結果 58
表4.2 人臉偵測方法之比較 58

[1] Ives, R.W.; Yingzi Du; Etter, D.M.; Welch, T.B., “A Multidisciplinary Approach to Biometrics”, IEEE, Volume 48, Issue 3, Aug. 2005 Page(s):462- 471.
[2] 施柏樑,運用特徵向量技術於手機取像之人臉辨識系統,國立中正大學電機工程研究所碩士論文
[3] Hamdy M. Kelash, Amany A. Kandeel, “Faces and Facial Features Detection in Color Images” Proceedings of the Geometric Modeling and Imaging― New Trends, 2006
[4] Montufar-Chaveznaza, R.; Gallardo, F.H.; Hernandez, S.P.; “Face Detection by Polling” IEEE Conference, Proceedings of the Intelligent Signal, 2005.
[5] Krishnan Nallaperumal, Ravi Subban, Krishnaveni, Lenin Fred, R.K.Selvakumar, “Human Face Detection in Color Images Using Skin Color and Template Matching Models for Multimedia on the Web” IEEE Conference, Proceedings of the Wireless and Optical Communications Networks, April 2006.
[6] Hyungkeun Jee, Kyunghee Lee, Sungbum Pan, “Eye and Face Detection using SVM” IEEE Conference, Proceedings of the 2004
[7] Rong Xiao, Ming-Jing Li, Hong-Jiang Zhang, “Robust Multipose Face Detection in Images” IEEE Transactions on Circuits and Systems for Video Technology, vol. 14, No. 1, January 2004.
[8] Y.H. Chan, “Face Detection System Based on Feature-Based Chrominance Colour Information” IEEE Conference, Proceedings of the International Conference on Computer Graphics Imaging and Visualization (CGIV’04), 2004.
[9] Rein-Lien Hsu, Anil K. Jain, “Face Detection in Color images” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, No.5, May 2002.
[10] Kah-Kay Sung, Tomaso Poggio, “Example-Based Learning for View-Based Human Face Detection” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 20, No.1, January 1998.
[11] V. Vezhnevets and V. Sazonov and A. Andreeva, "A Survey on Pixel-Based Skin Color Detection Techniques", Proc. Graphicon-2003, pp. 85-92, Moscow, Russia, September 2003.
[12] Sanjay Kr. Singh, D.S. Chauhan, Mayank Vatsa, Richa Singh*, “A Robust Skin Color Based Face Detection Algorithm,” Tamkang Journal of Science and Engineering, vol. 6, No. 4, pp. 227-234 (2003).
[13] Bo Hu, Qing Lin, Xuelei Kang and Guangmeng Chen, “A new algorithm for automatic white balance with priori,” the 2000 IEEE Asia-Pacific Conference on 4-6 Dec. 2000 Page(s):109 – 112.
[14] Son Lam Phung; Chai, D.; Bouzerdoum, A. “A Universal and Robust Human Skin Color Model using Neural Networks,” International Joint Conference on Neural Networks, vol.4, pp.2844-2849, 2001.
[15] D. Chai, A. Bouzerdoum, “A bayesian approach to skin color classification in YCbCr color space,＂IEEE Proceedings on TENCON 2000, Vol. 2, pp. 421-424, Sept. 2000.
[16] Sorian .M, and Martinkauppi.B, “Skin detection in video under changing illumination conditions,” IEEE Conference, Proceedings of the 12th international conference on image analysis and processing, 2003.
[17] F. Gasparini, R. Schettini, “Color Correction for Digital Photographs” IEEE Conference, Proceedings of the 12th international conference on image analysis and processing, 2003.
[18] Rafael C. Gonzalez, Richard E. Woods, “Digital Image Processing”, Second Edition.
[19] Haralick, Robert M., and Linda G. Shapiro, Computer and Robot Vision, Volume I, Addison-Wesley, 1992, pp. 28-48.
[20] 黃泰祥,具備人臉追蹤與辨識功能的一個智慧型數位監視系統,私立中原大學電子工程研究所碩士論文
[21] Birchfield, Stan, “An Elliptical Head Tracker”, 31st Asilomar Conference on Signals, Systems, and Computers, November 1997.