人臉偵測與追蹤之實作 蔡東昇、林國祥
E-mail: [email protected]
摘 要
本論文中，提出一個基於特徵點追蹤之人臉追蹤技術。本論文提出之人臉追蹤技術主要由兩個部份所組成：人臉偵測和人 臉追蹤。 首先，人臉偵測部分，使用Haar-Like特徵結合多分類器演算法的人臉偵測器，找出人臉的所在區域，為了移除 偵測錯誤的人臉區域，人眼資訊也被採用。完成人臉偵測後，本論文透過特徵擷取與特徵追蹤達成人臉追蹤。 為了評估所 提出的方法，藉由網路攝影機擷取各種不同的臉部動態影像。實驗結果顯示，本論文提出之人臉追蹤系統中，人臉偵測正 確檢測率高於96%，人臉追蹤正確檢測率高於91%，效果良好。實驗結果證明，在自然且具有雜訊環境下，本論文提出的 方法可以有效地達成人臉追蹤。
關鍵詞 : 人臉追蹤、人臉偵測、特徵追蹤、KLT追蹤、人眼偵測 目錄
封面內頁 簽名頁 中文摘要 iii ABSTRACT iv 誌謝 v 目錄 vi 圖目錄 viii 表目錄 x 第一章 緒論 1 1.1 研究動機 1 1.2 系統概要 1 1.3 人臉偵測相關技術 3 1.3.1 人臉追蹤困難之處 3 1.4 人臉追蹤相關技術 3 1.4.1 人臉追蹤困難之處 4 第二章 人臉偵測 5 2.1 人臉偵測之系統架構 5 2.2 人臉矯正 6 2.2.1 結構相似性指標 8 2.2.2 旋轉人臉影像 9 2.2.3 亮度調整 12 2.3 人眼偵測 14 2.3.1 局部二元樣本 16 2.3.2 支持向量機 17 第三章 人臉追蹤 20 3.1 人臉追蹤之系統架構 20 3.2 KLT特徵追蹤 22 3.3 重新人臉偵 測之程序 26 第四章 實驗結果 28 4.1 系統執行環境 28 4.2 人臉偵測之結果 29 4.2.1 Haar-Like特徵結合多分類器演算法的人 臉偵測之結果 29 4.2.2 人臉旋轉之結果 32 4.2.3 亮度調整與人眼偵測之結果 33 4.2.4 人臉偵測之結果 35 4.2.5 特徵擷取之結 果 40 4.2.6 人臉追蹤之結果 43 第五章 結論與未來研究方向 46 5.1 結論 46 5.2 未來研究方向 46 參考文獻 47
參考文獻
[1]P. Viola and M. Jones, “Robust real-time face detection,” International Journal of Computer Vision, vol. 57, no. 2, pp. 137-154, 2004.
[2]G. S. Lin, M. K. Chang, and Y. J. Chang, “Gender recognition based on multi-model information fusion,” APSIPA ASC 2011.
[3]Y. Andreu and R. A. Mollineda, “The role of face parts in gender recognition,” in Proc. LNCS Int’l Conf. on Image Analysis and Recognition, vol.5112, pp. 945-954, 2008.
[4]H. Fukai, H. Takimoto, Y. Mitsukura and M. Fukumi, “Age and Gender Estimation System based on Human Perception,” Proc. of 18th IEEE International Symposium on Robot and Human Interactive Communication, pp. 1143-1148, 2009.
[5]Olugbenga Ayinde, and Yee-Hong Yang, “Region-Based Face Detection,” Pattern Recognition, Vol. 35, pp. 2095-2017, 2002.
[6]S. Hayashi, and O. Hasegawa, “A Detection Technique for Degraded Face Images,” Conference on Computer Vision and Pattern Recognition, IEEE Computer Society, Vol. 2, pp. 1516-1512, 2006.
[7]L. Zhao, X. Sun, and X. Xu, “Face detection based on facial features,” Proceedings of the International Conference on Signal Processing, Vol. 3, Nov, pp. 16-20, 2006.
[8]M. Turkan, M. Pardes, and A.E. Cetin, “Edge Projections for eye localization,” Optical Engineering, Vol. 47, pp. 047007-1–047007-6, 2008.
[9]J. -S. Jang, and J. -H. Kim, “Fast and robust face detection using evolutionary pruning,” IEEE Transactions on Evolutionary Computation, Vol. 12, No. 5, pp. 562-571, 2008.
[10]P. S. Hiremath, and A. Danti, “Detection of Multiple Faces in an Image Using Skin Color Information and Lines-of-Separability Face Model,
” International Journal of Pattern Recognition and artificial Intelligence, Vol. 20, No. 1,pp. 39-61, 2006.
[11]Chin-Chung Han, Hong-Yuan Mark Liao, Kuo-Chung Yu, and Liang-Hua Chen, “Fast Face Detection Via Morphology-based Pre-Processing,” Pattern Recognition, Vol. 33, pp. 1701-1712, 2000.
[12]L. Goldmann, U. J. Monich, and T. Sikora, “Components and Their Topology for Detection in the Presence of Partial Occlusions,” IEEE Transactions on Information Forensics and Security, Vol. 2, No. 3, Sep, pp. 559–569, 2007.
[13]S. Bircheld, “Elliptical head tracking using intensity gradients and color histograms,” IEEE Conf. Computer Vision and Pattern Recognition, pp. 232–237, 1998.
[14]V. Kruger, R. Herpers, K. Daniilidis, and G. Sommer, “Teleconferencing using an attentive camera system,” Int. Conf. on Audio- and
Video-based Biometric Person Authentication, pp. 142–147, 1999.
[15]Y. Raja, J. McKenna, and S. Gong, “Tracking and segmenting people in varying lighting conditions using color,” Int. Conf. on Automatic Face- and Gesture- Recognition, pp. 228–233, 1998.
[16]Y. Zheng and Y. meng, “A swarm-intelligence based algorithm for face tracking,” Int. J. Intelligent Systems Technologies and Applications, vol. 7, no. 3, pp.266–281, 2009.
[17]Jeremiah R. Barr, Kevin W. Bowyer, and Patrick J. Flynn, “Detecting Questionable Observers Using Face Track Clustering,” 2011 IEEE Workshop on Applications of Computer Vision (WACV), pp. 182-189, 2011.
[18]Hao Ji, Fei Su, and Geng Du, “Multiple faces tracking based on joint kernel density estimation and robust feature descriptors,” 2009.
IC-NIDC 2009. IEEE International Conference on Network Infrastructure and Digital Content, pp. 680-685, 2009.
[19]Thanh Duc Ngo, Duy-Dinh Le, Shin’ichi Satoh, and Duc Anh Duong, “Robust Face Track Finding in Video Using Tracked Points,”
2008. SITIS '08. IEEE International Conference on Signal Image Technology and Internet Based Systems, pp. 59-64, 2008.
[20]C.-Y. Yu, Y.-C. Ouyang, C.-M. Wang, and C.-I Chang, “Adaptive inverse hyperbolic tangent algorithm for dynamic contrast adjustment in displaying scenes,” EURASIP Journal on Advanced in Signal Processing, vol. 2010, pp. 1-20, 2010.
[21]Z. Wang, A. C. Bovik, H. R. Sheikh and E. P. Simoncelli, “Image quality assessment: from error visibility to structural similarity,” IEEE Transactions on Image Processing, vol. 13, pp. 600-612, 2004.
[22]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, pp. 971-987, 2002.
[23]V. Vapnik, The Nature of Statisical Learning Theory, Springer-Verlag, New York, 1995.
[24]J. Shi, and C. Tomasi, “Good Feature to Track,” IEEE Conf. Computer Vision and Pattern Recognition, pp. 593-600, June 1994.
[25]T. Tommasini, A. Fueiello, and E. Trucco etal. , “Making good features track better,” Int’l Conf. Computer Vision and Pattern Recognition, pp. 178-183, 1998.
[26]Stan Birchfield, “Derivation of Kanade-Lucas-Tomasi Tracking Equation,” Unpublished, May 1996.
[27]C. Tomasi and T. Kanade, “Detection and Tracking of Point Features,” Carnegie Mellon University Technical Report CMU-CS-91-132, April 1991.
[28]C. Garcia and G. Tziritas, “Face Detection Using Quantized Skin Color Regions Merging and Wavelet Packet Analysis,” IEEE Transactions on Multimedia, vol. 1, pp. 264-277, 1999.
