研究生: |
林陳琦 Chen-Chi Lin |
---|---|
論文名稱: |
基於平均臉差異遮罩與二維離散餘弦轉換特徵之即時人臉性別辨識系統 Real-time face gender recognition system based on average face difference mask and features from 2D-DCT |
指導教授: |
郭景明
Jing-Ming Guo |
口試委員: |
蘇順豐
Shun-Feng Su 鍾聖倫 Sheng-Luen Chung 鍾國亮 Kuo-Liang Chung |
學位類別: |
碩士 Master |
系所名稱: |
電資學院 - 電機工程系 Department of Electrical Engineering |
論文出版年: | 2010 |
畢業學年度: | 99 |
語文別: | 中文 |
論文頁數: | 158 |
中文關鍵詞: | 人臉偵測 、平均臉差異 、性別辨識 、支持向量機 |
外文關鍵詞: | face detection, average face difference, gender recognition, SVM |
相關次數: | 點閱:420 下載:4 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
早期的全自動人臉性別辨識系統受限於不成熟的人臉偵測技術。近年來人臉偵測技術逐漸地成熟,因此人臉性別辨識也開始受到重視,並可應用於廣告和監控領域中。
在人臉偵測方面,若整個系統只使用同一種特徵系列將造成偵測率與錯誤率受到該特徵的限制,因此本論文使用三個完全不同特性的分類器串聯而成,第一、二層使用基於統計的人臉遮罩和區塊總和濾除非人臉,第三層使用基於矩形特徵的AdaBoost強分類器,第一層的任務是迅速排除大量非人臉區塊,使整體處理速度能達到即時處理。偵測到人臉後即可進行人臉性別辨識。
在人臉性別辨識方面,前人的方法只單獨考慮空間域或頻率域的分佈,但人臉偵測器取得的人臉同時包含一些空間域與頻率域的非人臉雜訊。有鑑於前人方法所具有的問題,本論文提出一個機制用以刪除或降低空間域與頻率域中的非人臉雜訊,刪除這些對人臉性別辨識無幫助的資訊將能提高識別率與處理效率。本論文首先提出基於平均人臉差異的性別遮罩,其目的是為了刪除空間域中包含的一些干擾資訊或非人臉雜訊。接著使用二維離散餘弦轉換(2D-DCT)刪除高頻雜訊,如此能保留適當低頻能量形成較小維度且包含較多資訊的性別特徵向量。最後使用支持向量機(SVM)辨識所得的性別特徵向量。
為了能即時處理,本論文分析整個系統所有階段的處理速度最佳化策略。在人臉性別辨識的實驗階段,本論文使用FERET人臉資料庫進行識別率、處理速度和訓練時間的比較,比較的方法包含提出的方法與四種前人提出的方法,實驗結果顯示提出的方法識別率最高、訓練時間最短和介於中間的處理速度。
最後本論文使用PTZ攝影機、USB影像擷取卡、筆記型電腦和RS232傳輸線完成一套可用於監控的自動人臉性別辨識系統,此系統可使用介面化程式下達控制命令,並經由RS232傳輸線控制PTZ攝影機的許多功能,包括開關機、旋轉、縮放和背光補償等。
Early full-automatic face gender recognition systems are quite limited since they were just developed from the original face detection technology. However the face detection techniques have been gradually getting precise and reliable in recent years, so other advanced trends, such as face gender recognition has become an interesting topic. Some of the most practical applications of face gender recognition are in public advertisement and surveillance.
In face detection phase, if the whole system uses only one feature series, the detection rate is limited as it is only determined by this feature series. To overcome this problem, this thesis discusses connecting three classifiers of totally different features in series. The first two classifying layers use statistics-based face mask, and the sum of the block filters out non-face features. The third layer uses rectangle-feature-based AdaBoost strong classifier. The objective of the first layer is that fast filtering a large amount of non-face features, which enhances the overall processing speed to reach real-time level. The face gender detection phase is then carried on after the system successfully detects faces.
In face gender recognition phase, previous methods only consider the distribution of the space domain or frequency domain. However, the obtained face from the detector includes non-face noise of some spatial domain and frequency domain at the same time. This paper proposes a mechanism to delete or reduce non-face noise in the spatial domain and frequency domain. Deleting the information that are ineffective or even causing noise to face gender recognition can improve recognition rate and precision. The proposed scheme first employ average-face-difference-based gender mask. Its purpose is to delete some interference information or non-face noise included in the spatial domain. The next step is to apply the 2D-DCT to truncate high-frequency noise, thus the energy is kept at appropriate low frequency. The gender feature vector not only has smaller dimension but also includes more useful information. Finally, the gender is classified with SVM using gender feature vector.
To achieve real-time processing speed, this paper analyzes the speed optimization tactics at all stages. In the experiments, the FERET face database is employed for performance testing, including recognition rate, processing speeds and training time. The methods for comparisons include the proposed method and the four former proposed methods. Experimental result shows that the proposed method provides the best recognition rate with shortest training time, while keeping the similar processing speed.
The equipments used in this research including PTZ camera, USB video capture card, notebook and RS232 transmission. Those equipments are enough to make a set of automatic face gender recognition system to be used in surveillance. This system can use interface programming to control a lot of functions of PTZ camera via RS232 transmission line, including power, rotating, scale and Backlight compensation.
[01]M. Turk and A. Pentland, "Eigenfaces for recognition," Journal of Cognitive Neuroscience, vol. 3, no. 1, pp. 71-86, Jan. 1991.
[02]E. Hjelmas and B. K. Low, "Face detection: a survey," Computer Vision and Image Understanding, vol. 83, no. 3, pp. 236-274, Sep. 2001.
[03]L. Meng and T. Q. Nguyen, "Two subspace methods to discriminate faces and clutters," IEEE International Conference on Image Processing (ICIP), vol. 2, pp. 215-218, Sep. 2000.
[04]E. Osuna, R. Freund and F. Girosi, "Training support vector machines: an application to face detection," IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), pp. 130-136, Jun. 1997.
[05]B. E. Boser, I. M. Guyon and V. N. Vapnik, "A training algorithm for optimal margin classifiers," Fifth Annual ACM Workshop on Computational Learning Theory, pp. 144-152, July 1992.
[06]C. Cortes and V. Vapnik, "Support-vector networks," Machine Learning, vol. 20, no. 3, pp. 273-297,Sep. 1995.
[07]V. P. Kumar and T. Poggio, "Learning-based approach to real time tracking and analysis of faces," Fourth IEEE International Conference on Automatic Face and Gesture Recognition, pp. 96-101, Mar. 2000.
[08]J. Ng and S. Gong, "Multi-view face detection and pose estimation using a composite support vector machine across the view sphere," IEEE International Workshop on Recognition, Analysis, and Tracking of Faces and Gestures in Real-Time Systems, pp. 14-21, Sep. 1999.
[09]H. A. Rowley, S. Baluja and T. Kanade, "Neural network-based face detection," IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), vol. 20, no. 1, pp. 23-38, Jan. 1998
[10]H. A. Rowley, S. Baluja and T. Kanade, "Rotation invariant neural network-based face detection," IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), pp. 963, Jun. 1998
[11]R. Feraud, O. J. Bernier, J. E. Viallet and M. Collobert, "A fast and accurate face detector based on neural networks," IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), vol. 23, no.1, pp. 42-53, Jan. 2001.
[12]M. A. Kramer, "Nonlinear principal component analysis using autoassociative neural networks," American Institute of Chemical Engineers Journal, vol. 37, no. 2, pp. 233-243, Feb. 1991.
[13]P. Viola and M. J. Jones, "Robust real-time face detection," International Journal of Computer Vision, vol. 57, no. 2, pp. 137-154, May 2004.
[14]Y.-W. Wu and X.-Y. Ai, "An improvement of face detection using AdaBoost with color information," ISECS International Colloquium on Computing, Communication, Control, and Management (CCCM), vol. 1, pp. 317-321, Aug. 2008.
[15]A. Hadid, M. Pietikainen and T. Ahonen, "A discriminative feature space for detecting and recognizing faces," IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), vol. 2, pp. 797-804, July 2004.
[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 (TPAMI), vol. 24, no. 7, pp. 971-987, July 2002.
[17]B. Moghaddam and M. H. Yang, "Learning gender with support faces," IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), Vol. 24, no. 5, pp. 707-711, May 2002.
[18]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.
[19]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.
[20]P. Viola and M. J. Jones, "Robust real-time object detection," Second International Workshop on Statistical and Computational Theories of Vision, July 2001.
[21]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.
[22]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.
[23]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.
[24]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.
[25]H.-Z. Ai, X.-P. Xiao and G.-Y. Xu, "Face detection and retrieval," Chinese Journal of Computers, vol. 26, no. 7, pp. 874-881, July 2003.
[26]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.
[27]S. Baluja and H. A. Rowley, "Boosting sex identification performance," International Journal of Computer Vision, vol. 71, no. 1, pp. 111-119, Jan. 2007.
[28]H. Lu and H. Lin, "Gender recognition using AdaBoosted feature," Third International Conference on Natural Computation, Vol. 2, pp. 646-650, Aug. 2007.
[29]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.
[30]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.
[31]E. Makinen and R. Raisamo, "An experimental comparison of gender classification methods," Pattern Recognition Letters, vol. 29, no. 10, pp. 1544-1556, July 2008.
[32]H. Lu, Y. Huang, Y. Chen and D. Yang, "Automatic gender recognition based on pixel-pattern-based texture feature," Journal of Real-Time Image Processing, vol. 3, no. 1-2, pp. 109-116, Mar. 2008.
[33]B. C. Shen, C. S. Chen and H. H. Hsu, "Fast gender recognition by using a shared-integral-image approach," IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 521-524, April 2009.
[34]R. E. Schapire and Y. Singer, "Improved boosting algorithms using confidence-rated predictions," Machine Learning, vol. 37, no. 3, pp. 297–336, Dec. 1999.
[35]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.
[36]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.
[37]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.
[38]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.
[39]趙楠,「基於AdaBoost算法的人臉偵測」,北京大學,北京,2005。
[40]吳旻峰,「基於像素導向之階層式特徵與統計式遮罩AdaBoost人臉偵測」,碩士論文,國立台灣科技大學,台北,2009。
[41]V. Vapnik, "Estimation of dependences based on empirical data," Springer-Verlag New York, 1982.
[42]E. Osuna, R. Freund and F. Girosi, "An improved training algorithm for support vector machines," IEEE Workshop Neural Networks for Signal Processing, pp. 276-285, Sep. 1997.
[43]J. C. Platt, "Fast training of support vector machines using sequential minimal optimization," Advances in Kernel Methods - Support Vector Learning, pp. 185-208, 1999.
[44]C.-C. Chang and C.-J. Lin, "LIBSVM : a library for support vector machines," 2001. Software available at http://www.csie.ntu.edu.tw/~cjlin/libsvm
[45]J. M. Guo, C. C. Lin and H. S. Nguyen, "Face gender recognition using improved appearance-based average face difference and support vector machine," International Conference on System Science and Engineering (ICSSE), pp. 637-640, July 2010.
[46]MIT-CBCL face database available at http://cbcl.mit.edu/cbcl/software-datasets/FaceData2.html
[47]P. J. Phillips, H. Wechsler, J. Huang and P. J. Rauss, "The FERET database and evaluation procedure for face recognition algorithms," Image and Vision Computing Journal, vol. 16, no. 5, pp. 295-306, April 1998.
[48]P. J. Phillips, H. Moon, S. A. Rizvi and P. J. Rauss, "The FERET evaluation methodology for face recognition algorithms," IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) , vol. 22, no. 10, pp. 1090-1104, Oct. 2000.
[49]CMU frontal face images available at http://vasc.ri.cmu.edu//idb/html/face/frontal_images/index.html
[50]SONY EVI-D70技術手冊,http://pro.sony.com/bbsc/ssr/product-EVID70P/