進階搜尋


下載電子全文  
系統識別號 U0026-2307201016210700
論文名稱(中文) 基於人臉特徵的性別辨識系統之實現
論文名稱(英文) An Implementation of Gender Recognition System Based on Facial Features
校院名稱 成功大學
系所名稱(中) 工程科學系碩博士班
系所名稱(英) Department of Engineering Science
學年度 98
學期 2
出版年 99
研究生(中文) 何信承
研究生(英文) Hsin-Cheng Ho
學號 n9697114
學位類別 碩士
語文別 中文
論文頁數 77頁
口試委員 指導教授-王明習
口試委員-孫永年
口試委員-鄭國順
口試委員-鄭承昌
中文關鍵字 性別辨識  辨別能力分析  二維主成分分析  興趣運算元 
英文關鍵字 Gender recognition  Discrimination Power Analysis (DPA)  2-D Principal Component Analysis (2DPCA)  Interest Operator (IO) 
學科別分類
中文摘要 隨著資訊科技之演進,以臉部特徵來辨識性別之人機互動相關應用,逐漸變成一個相當受到重視的議題。本論文提出一個以臉部特徵來做性別辨識之系統,首先要能偵測取得之影像是否有臉部存在,故對取得之影像利用物件偵測演算法對影像進行人臉部位偵測並定位,接著將切割出的臉部影像做正規化。在特徵擷取方面,我們考慮三種特徵分析方式來選用特徵值,分別為辨別能力分析(Discrimination Power Analysis, DPA)、二維主成分分析(2-D principal component analysis, 2DPCA)及興趣運算元(Interest Operator, IO)。並針對選出之每一種特徵分別測試其成效,同時也將三種特徵結合來測試,最後利用多數決之投票方式來決定性別。在台灣近視人口的比例相當高,根據研究,人臉影像中眼鏡的存在對特徵擷取會有不良影響,本研究針對此問題,提出了眼鏡偵測與消除方法以減少因眼鏡存在,造成對相關特徵之影響。由實驗結果可以得知本研究提出的性別辨識方法具有不錯的效能。
英文摘要 Gender recognition based on facial features is an attractive issue in the recent years. In this thesis, a gender recognition system was proposed. For an input image, the facial region is firstly segmented by face detection algorithm. The segmented images are normalized to a uniform size. Three methods, Discrimination Power Analysis (DPA), 2-D Principal Component Analysis (2DPCA) and Interest Operator (IO) are considered to extract the features of the normalized face image. The DPA is used to select the coefficients from the DCT transformed image of the face image which provide more discriminating capability. The 2DPCA is based on the image matrix, and it is simpler to use for image feature extraction. The IO calculates variation information in different directions of the image pixel intensity. In order to evaluate the affect of the recognition rate for these images with eye glasses, the proposed method considered two methods to erase the eye glasses and their results are compared. To compare the effectiveness of the features extracted from the three ways, different combination ways for these features are considered and the results are compared. Finally, the majority voting method is applied to get the results for the features extracted from all three methods. From the experimental results, it is shown that the proposed method can perform well.
論文目次 摘要 i
Abstract ii
誌謝 iii
目錄 iv
表目錄 vi
圖目錄 vii
第一章 緒論 1
1.1 研究動機與目的 1
1.2 相關研究 3
1.3 論文架構 11
第二章 背景知識 13
2.1 人臉部位偵測 13
2.1.1 矩形特徵 13
2.1.2 積分影像 14
2.1.3 Adaboost演算法 17
2.1.4 瀑布型分類器 19
2.2 人臉特徵擷取 20
2.2.1 辨別能力分析 21
2.2.2 二維主成分分析 25
2.2.3 興趣運算元 27
2.3 支援向量機架構 32
2.3.1 線性可分離支援向量機 33
2.3.2 線性不可分離支援向量機 36
2.3.3 非線性支援向量機 39
第三章 人臉影像之性別辨識 40
3.1 前處理與人臉部位偵測 41
3.2 影像正規化 47
3.3 眼鏡偵測與消除 50
3.3.1 眼鏡偵測 50
3.3.2 鏡框消除 53
3.4 人臉特徵擷取 55
3.5 性別分類 59
第四章 實驗結果與討論 61
4.1 實驗環境 61
4.2 實驗結果 62
4.2.1 選擇特徵維度測試 62
4.2.2 系統測試 64
第五章 結論與未來研究方向 73
5.1 結論 73
5.2 未來研究方向 74
參考文獻 76

參考文獻 [1] Mäkinen, E. and R. Raisamo, "An experimental comparison of gender classification methods," Pattern Recognition Letters, vol. 29, pp. 1544-1556, 2008.
[2] Rein-Lien Hsu, Abdel-Mottaleb, M., Jain, A.K., "Face detection in color images," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol.24, no.5, pp. 696-706, 2002.
[3] Sobottka, K. and I. Pitas, "Extraction of facial regions and features using color and shape information," Proceedings of the 13th International Conference on Pattern Recognition, Vienna , Austria, 25-29 Aug. 1996.
[4] Jin, Zhong, Lou, Zhen, Yang, Jingyu, Sun, Quansen, "Face detection using template matching and skin-color information," Neurocomputing, vol.70, pp. 794-800, 2007.
[5] Turk, M. A. and A. P. Pentland, "Face recognition using eigenfaces," IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Maui, Hawaii, USA, 3-6 Jun. 1991.
[6] Belhumeur, P., J. Hespanha, David J. Kriegman, "Eigenfaces vs. Fisherfaces: Recognition using class specific linear projection," IEEE Transactions On Pattern Analysis And Machine Intelligence, Vol. 19, No. 7, pp. 711-720, 1997.
[7] Viola, P. and M. J. Jones, "Robust Real-Time Face Detection," International Journal of Computer Vision, vol. 57, pp. 137-154, 2004.
[8] Viola, P., M. J. Jones, Snow, Daniel, "Detecting Pedestrians Using Patterns of Motion and Appearance," International Journal of Computer Vision, vol. 63, pp. 153-161, 2005.
[9] Dabbaghchian, S., M. P. Ghaemmaghami, Aghagolzadeh, Ali, "Feature extraction using discrete cosine transform and discrimination power analysis with a face recognition technology," Pattern Recognition, vol. 43, pp. 1431-1440, 2010.
[10] Xu, Yong, Yao, Lu, Zhang, David, Yang, Jing-Yu, "Improving the interest operator for face recognition," Expert Systems with Applications, vol. 36, pp. 9719-9728, 2009.
[11] Wang, Chuan Xu, Liu, Yun, Li, Zuo-yong, "Algorithm Research of Face Image Gender Classification Based on 2-D Gabor Wavelet Transform and SVM," International Symposium on Computer Science and Computational Technology, Shanghai, 20-22 Dec. 2008.
[12] Lian, H.-C. and B.-L. Lu, "Multi-view gender classification using multi-resolution local binary patterns and support vector machines," International Journal of Neural Systems, vol. 17, pp. 479-487, 2007.
[13] Hart, P., "Nearest neighbor pattern classification," IEEE Transactions on Information Theory, vol. 13, pp. 21-27, 1967.
[14] Freund, Y. and R. Schapire, "A desicion-theoretic generalization of on-line learning and an application to boosting," Journal of Computer and System Sciences, vol. 55, pp. 23-37, 1997.
[15] Cortes, C. and V. Vapnik, "Support-vector networks," Machine Learning, vol.20, pp. 273-297, 1995.
[16] Boser, B. E., I. M. Guyon, Vladimir N. Vapnik, "A training algorithm for optimal margin classifiers," Proceedings of the fifth annual workshop on Computational learning theory, Pittsburgh, Pennsylvania, United States, 27-29 July 1992.
[17] Li Lu and Pengfei Shi, "Fusion of multiple facial regions for expression-invariant gender classification," IEICE Electron. Express, vol. 6, No. 10, pp. 587-593, 2009.
[18] Jian, Yang, Zhang, D., Frangi, A. F., Jing-yu, Yang, "Two-dimensional PCA: a new approach to appearance-based face representation and recognition," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 26, pp. 131-137, 2004.
[19] Ueki, K., Komatsu, H., Imaizumi, S., Kaneko, K., Sekine, N., Katto, J., Kobayashi, T., "A method of gender classification by integrating facial, hairstyle, and clothing images," Proceedings of the 17th International Conference on Pattern Recognition, Cambridge, United Kingdom, 23-26 Aug. 2004.
[20] Alexandre, L.A. "Gender recognition: A multiscale decision fusion approach," Pattern Recognition Letters (2010), doi:10.1016/j.patrec.2010.02.010
[21] Saito, Y., Y. Kenmochi, Kotani, K., "Estimation of eyeglassless facial images using principal component analysis," International Conference on Image Processing, Kobe, Japan, 25-28 Oct. 1999.
[22] Chang, C. C., Lin, C. J., "LIBSVM: a Library for Support Vector Machines," [Online] http://www.csie.ntu.edu.tw/~cjlin/papers/libsvm.pdf
[23] Wen Gao, Bo Cao, Shiguang Shan, Xilin Chen, Delong Zhou, Xiaohua Zhang, Debin Zhao. "The CAS-PEAL Large-Scale Chinese Face Database and Baseline Evaluations," IEEE Trans. on System Man, and Cybernetics (Part A), vol.38, pp. 149-161, 2008.
論文全文使用權限
  • 同意授權校內瀏覽/列印電子全文服務,於2011-07-29起公開。
  • 同意授權校外瀏覽/列印電子全文服務,於2011-07-29起公開。


  • 如您有疑問,請聯絡圖書館
    聯絡電話:(06)2757575#65773
    聯絡E-mail:etds@email.ncku.edu.tw