基于综合集成的人脸识别
|
摘要
人脸识别在模式识别领域的发展和应用方面都有着重要意义，目前是一个非常活跃的研究方向。人脸识别问题一般可以描述为：从背景图像中检测人脸；抽取人脸识别特征；最后进行匹配和识别。本文在人脸识别的三个主要环节上均进行了研究工作，主要工作体现在以下几个方面：
(1)提出了基于子空间特征的人脸检测方法。论文中利用K-L变换得到特征脸子空间，进而得到样本在该空间上的投影系数。在此基础上，采用模糊聚类和加权K近邻分类器相结合的方法进行人脸检测，使得人脸检测速度明显提高。
(2)在基于子空间方法的人脸识别中，为了克服NMF方法不能提取图像局部特征信息的缺点，提出用局部NMF方法提取人脸子空间特征，该方法能够提取图像的局部特征信息，有利于提高人脸识别率。将Bagging思想用于神经网络，进一步提高神经网络的分类准确率和泛化能力。
(3)在基于几何特征的人脸识别中，为了克服眼睛定位不准确的缺点，提出一种根据几何特征定位人眼，然后采用特征眼模板进行二级验证的定位方法。
(4)考虑到子空间特征和几何特征反映的是不同模式的特征信息，提出综合集成的人脸识别方法。在综合集成过程中，为克服分类器权系数确定的不灵活性，提出基于神经网络和遗传算法动态确定权系数估计器的方法。
实验证明，基于以上方法的人脸识别，比单独采用子空间特征或几何特征进行识别，识别率明显提高。
Face recognition plays an important role in the development and application of pattern recognition. At present, it is an active research topic. A general statement of the problem can be formulated as three steps: face detection, facial feature extraction, match and recognition. This paper deals with all above three steps, most work is described as follows:
(1) We present a method of face detection based on subspace feature. In this paper, K-L transformation is used to get subspace of eigenfaces. Then projection coefficients on the subspace can be obtained. Based on this, blur cluster and weighted k-nearest neighbor algorithm are adopted to detect faces, which greatly enhances the speed of face detection.
(2) In face recognition based on subspace feature, we propose local NMF for extraction of subspace of eigenfaces in order to solve the problem that NMF can't extract local feature. Our method can fully extract local feature of image, which is helpful in enhancing recognition rate. Meanwhile, Bagging algorithm is used, which improves classification accuracy and generalization of neural network.
(3) In face recognition based on geometrical feature, we present a method of localization in order to overcome the problem that eyes can't be located accurately. Firstly we localize eyes according to geometrical feature, then we use eigeneyes template to validate the result.
(4) We present a metasynthesis method of face recognition, considering that subspace and geometrical feature reflect different feature information. In metasynthesis, we present a method of deciding dynamically estimator of weights based on neural network and genetic algorithm, in order to overcome the problem that weights are decided unskillfully.
The experimental results prove that recognition rate is higher based
on our methods than that based on individual geometrical feature or individual subspace feature.
(1)提出了基于子空间特征的人脸检测方法。论文中利用K-L变换得到特征脸子空间，进而得到样本在该空间上的投影系数。在此基础上，采用模糊聚类和加权K近邻分类器相结合的方法进行人脸检测，使得人脸检测速度明显提高。
(2)在基于子空间方法的人脸识别中，为了克服NMF方法不能提取图像局部特征信息的缺点，提出用局部NMF方法提取人脸子空间特征，该方法能够提取图像的局部特征信息，有利于提高人脸识别率。将Bagging思想用于神经网络，进一步提高神经网络的分类准确率和泛化能力。
(3)在基于几何特征的人脸识别中，为了克服眼睛定位不准确的缺点，提出一种根据几何特征定位人眼，然后采用特征眼模板进行二级验证的定位方法。
(4)考虑到子空间特征和几何特征反映的是不同模式的特征信息，提出综合集成的人脸识别方法。在综合集成过程中，为克服分类器权系数确定的不灵活性，提出基于神经网络和遗传算法动态确定权系数估计器的方法。
实验证明，基于以上方法的人脸识别，比单独采用子空间特征或几何特征进行识别，识别率明显提高。
Face recognition plays an important role in the development and application of pattern recognition. At present, it is an active research topic. A general statement of the problem can be formulated as three steps: face detection, facial feature extraction, match and recognition. This paper deals with all above three steps, most work is described as follows:
(1) We present a method of face detection based on subspace feature. In this paper, K-L transformation is used to get subspace of eigenfaces. Then projection coefficients on the subspace can be obtained. Based on this, blur cluster and weighted k-nearest neighbor algorithm are adopted to detect faces, which greatly enhances the speed of face detection.
(2) In face recognition based on subspace feature, we propose local NMF for extraction of subspace of eigenfaces in order to solve the problem that NMF can't extract local feature. Our method can fully extract local feature of image, which is helpful in enhancing recognition rate. Meanwhile, Bagging algorithm is used, which improves classification accuracy and generalization of neural network.
(3) In face recognition based on geometrical feature, we present a method of localization in order to overcome the problem that eyes can't be located accurately. Firstly we localize eyes according to geometrical feature, then we use eigeneyes template to validate the result.
(4) We present a metasynthesis method of face recognition, considering that subspace and geometrical feature reflect different feature information. In metasynthesis, we present a method of deciding dynamically estimator of weights based on neural network and genetic algorithm, in order to overcome the problem that weights are decided unskillfully.
The experimental results prove that recognition rate is higher based
on our methods than that based on individual geometrical feature or individual subspace feature.
引文
[1] 张翠平，苏光大．人脸识别技术综述．中国图像图形学报，2001，5(11)：885～894
[2] 梁路宏，艾海舟，徐光等．人脸检测研究综述．计算机学报，2002，25(5)：449～457
[3] Sung K, Poggio T. Example-based learning for view based human face detection. IEEE Trans Pattern Analysis and Machine Intelligence, 1998, 20(1): 39～51
[4] Dietterich TG. An experimental comparison of three methods for constructing ensembles of decision trees: bagging, boosting, and randomization. Machine Learning, 2000, 40(2): 139-157
[5] 周激流，张晔．人脸识别理论研究进展．计算机辅助设计与图形学学报，1999，11(2)：180～184
[6] C. H. Lee, J. S. Kim, K. H. Park, Automatic human face location in a complex background, Pattern Recognition 1996, 29: 1877～1889
[7] H. Rowley, S. Baluja, T. Kanade. Neural network-based face detection. IEEE Trransactions on Pattern Analysis and Machine Intelligence, 1998, 20(1): 23～38
[8] Chellappa R, Wilson C L, Sirohey S. Human and machine recognition of faces: A Survey. Proceedings of the IEEE, 1995, 83(5): 705～740
[9] Turk M, Pentland A. Face recognition using eigenfaces. In: Proc. of IEEE Conf. on CVPR, 1991: 586～591
[10] 严超，苏光大．人脸特征的定位与提取．中国图像图形学报，1998，3(5)：375～379
[11] Berto R, Poggio T. Face recognition: Feature versus templates. IEEE Trans. on PAMI, 1993, 15(10): 1042～1052.
[12] Ziquan Hong. Algebraic feature extraction of image for recognition. Pattern Recognition, 1991, 24(3): 211～219
[13] Brunelli R, Poggio T. Face recognition: feature versus template. IEEE Tans. On PAMI 1993, 15(10): 1042～1052
[14] M. Turk and A. Pentland. Eigenfaces for recognition. Journal of Neuroscience, 1991, 3(1): 71～86
[15] Jun Zhang, Yong Yan, Martin Lades. Face recognition: Eigenface, Elastic Matching, and Neural Nets. Proceedings of the IEEE, 1998, 85(9): 1422～1435
[16] 田原，吴更石，梁德群．计算机人脸识别的研究现状与进展．无线电工程，1997，27(3)：5～14
[17] Cheng Y Q, Zhuang Y M, Wang H F, et al. An efficient recognition approach to human faces. ACTA Automatica Sinica, 1993, 19(1): 54～61
[18] Zhou Jiliu. A study on automatic recognition system for human profite. Journal of Sichuan University(Natural Science Edition), 1994, 31(1):100～104
[19] Samal A, lyengar P A. Automatic recognition and analysis for human faces and facial expressions: A survey. Pattern Recognition, 1992, 25(2): 65～77
[20] Lee S Y, Ham Y K, Park R H. Recognition of human front faces using knowledge-based feature extraction and neuro fuzzy algorithm. Pattern Recognition, 1996, 29(11): 1863～1876
[21] Huang C L, Chen C W. Human facial feature extraction for face interpretation and recognition. Pattern Recognition, 1992, 25(12): 1435～1444
[22] H. A. Rowley, S. Baluja, T. Kanade, Neural network-based face detection, IEEE Trans. PAMI, 1998, 20(1): 23～38
[23] K. K. Sung, T. Poggio, Example based learning for view-based human face detection, IEEE Trans. PAMI, 1998, 20(1): 39～50
[24] Peng Z, Too L, Xu G, et al. Detecting facial feature on images with multiple faces. In: Proc. conf. on Multimodal Interfaces, Beijing, China, 2000:191～198
[25] Tao L. M. Color perception for machine vision and human skin-color features: Ph. D thesis. Beijing: Tsinghua University. 2001. (in Chinese)
[26] Hong W, Ai H, Zhuang L, et al. Video based face verification.
In: Proc. of SPIE, 2001, 45(50):171～175
[27] Liang L, Ai H, He K, et al. Face detection based on the matching of multiple related templates, Chinese Journal of Software, 12(1): 94～102
[28] 边肇祺，张学工．模式识别．北京：清华大学出版社，2000：212～228
[29] Rowley, HA, S. Baluja, T. Neural Network-Based Face Detection, IEEE Transactions on PAMI, 1998, 20(1):23～38
[30] LIANG, Haizhou AI, Kezhong HE, et al. Multiple Related Template Matching Based Face Detection. Chinese Journal of Software, 2001, 12(1): 94～102.
[31] P. Viola & M. Jones, Rapid Object Detection using a Boosted Cascade of Simple Features, Proc. of IEEE Conf. on Computer Vision and Pattern Recognition, 2001, 12(1):511～518
[32] D. Graupe, Principles of Artificial Neural Networks, World Scientific Publishing Co. 1997, 15(3): 191～222
[33] R. Feraud, O. J. Bernier, J. E. Viallet, M. Collobert. A fast and accurate face detector based on neural networks, IEEE Trans. PAMI, 2001, 23(1):42～53
[34] L. Breiman. Bagging predictors. Machine Learning, 1996, 26(2): 123～140
[35] P. Buhlmann and B. Yu. Analyzing bagging. Ann. of Statist, 2002, 30(6):927～961
[36] S. X. Chen and P. Hall. Effects of bagging and bias correction on estimators defined by estimating equations. Statistica Sinica, 2003, 13(30): 97～109
[37] Patrik O. Hoyer. Non-negative Sparse Coding. Nin Proc. IEEE Workshop on Neural Networks for Signal Processing. 2002, 7(5): 557～565
[38] Qi X F, Palmieri F. Theoretical analysis of evolutionary algorithms with an infinite population size in continuous space, Part Ⅱ: Basic properties of selection and mutation. IEEE Trans on Neural Networks, 1994, 5(1): 120～129
[39] D. D. Lee, H. S. Seung. Algorithms for non-negative matrix
factorization. In Advances in Neural Information Processing Systems MIT Press, 2001, 13(7): 556～562
[40] 樊重俊，王浣尘．遗传算法的改进与应用．上海交通大学学报，1998，32(12)：127～131
[41] 夏利民，谷士文，宁国英．基于小波包分析的货车滚动轴承故障诊断．中国铁道科学，2002，23(6)：60～64
[42] 张俊，何昕，李介谷．基于面部几何特征点提取的人脸识别方法．红外与激光工程，1999，28(4)：40～44
[43] 范宏深，倪国强，申会堂．利用几何特性及神经网络进行人脸探测技术的研究．光学技术，2002，28(2)：105～108
[44] 梁路宏，艾海舟，何克忠．基于多模板匹配的单人脸检测．中国图像图形学报，1999，4(A)(10)：825～829
[45] 刘小军．人脸识别技术研究：[博士学位论文]．北京：中国科学院电子学研究所，2001，7
[46] Schapire, R. E., Freund, Y., Bartlett, P. et al. Boosting the margin: A new explanation for the effectiveness of voting methods. In Machine Learning: Proceedings of Fourteenth International Conference, 1997, 25(2): 322～330
[47] W. Zhao, R. Chellappa, A. Krishnaswamy. Discriminant analysis of principal components for face recognition. In International Conference on Automatic Face and Gesture Recognition 1998, 4(1): 336～341
[48] Alex Pentland, Baback Moghaddam, rhad Starner. View-Based and Modular Eigenspaces for Face Recognition. IEEE Conf. on Computer Vision and Pattern Recognition, MIT Media Laboratory Tech. Report 1994, 26(6): 245～249
[49] Breiman L. Bagging predictors. Machine Learning, 1996, 26(2): 123～140
[50] S. D. Bay. Nearest Neighbor Classification from Multiple Feature Subsets. Intelligent Data Analysis, 1999, 3(2): 191～209
[51] D. Lee, H. Seung. Learning the parts of objects by non-negative matrix factorization. Natur, 1999, 401(3): 788～791
[52] Thierry Denoeux. A k-Nearest Neighbor Classification Rule
Based on Dempster2Shafer Theory[J]. IEEE Trans. on Systems Man and Cybernetics, 1995, 25(5): 804～813
[53] Aleix M, Mart nez, and Avinash C. Kak. PCA versus LDA. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001, 23(2): 228～233
[54] Bauer, E, Kohavi R. An Empirical Comparison of Voting Classification Algorithms: Bagging, Boosting, and Variants. Machine Learning, 1999, 36(3): 105～139
[55] R. Feraud, O. J. Bernier. A fast and accurate face detection based on neural network. IEEE PAMI, 2001, 23(1): 42～53
[56] G. Donato, M. S. Bartlett, J. C. Hager, P. Ekman, et al. Sejnowski, "Classifying Facial Actions," IEEE Transactions on Pattern Analysis and Machine Intelligence, 1999, 21(2): 974-989
[2] 梁路宏，艾海舟，徐光等．人脸检测研究综述．计算机学报，2002，25(5)：449～457
[3] Sung K, Poggio T. Example-based learning for view based human face detection. IEEE Trans Pattern Analysis and Machine Intelligence, 1998, 20(1): 39～51
[4] Dietterich TG. An experimental comparison of three methods for constructing ensembles of decision trees: bagging, boosting, and randomization. Machine Learning, 2000, 40(2): 139-157
[5] 周激流，张晔．人脸识别理论研究进展．计算机辅助设计与图形学学报，1999，11(2)：180～184
[6] C. H. Lee, J. S. Kim, K. H. Park, Automatic human face location in a complex background, Pattern Recognition 1996, 29: 1877～1889
[7] H. Rowley, S. Baluja, T. Kanade. Neural network-based face detection. IEEE Trransactions on Pattern Analysis and Machine Intelligence, 1998, 20(1): 23～38
[8] Chellappa R, Wilson C L, Sirohey S. Human and machine recognition of faces: A Survey. Proceedings of the IEEE, 1995, 83(5): 705～740
[9] Turk M, Pentland A. Face recognition using eigenfaces. In: Proc. of IEEE Conf. on CVPR, 1991: 586～591
[10] 严超，苏光大．人脸特征的定位与提取．中国图像图形学报，1998，3(5)：375～379
[11] Berto R, Poggio T. Face recognition: Feature versus templates. IEEE Trans. on PAMI, 1993, 15(10): 1042～1052.
[12] Ziquan Hong. Algebraic feature extraction of image for recognition. Pattern Recognition, 1991, 24(3): 211～219
[13] Brunelli R, Poggio T. Face recognition: feature versus template. IEEE Tans. On PAMI 1993, 15(10): 1042～1052
[14] M. Turk and A. Pentland. Eigenfaces for recognition. Journal of Neuroscience, 1991, 3(1): 71～86
[15] Jun Zhang, Yong Yan, Martin Lades. Face recognition: Eigenface, Elastic Matching, and Neural Nets. Proceedings of the IEEE, 1998, 85(9): 1422～1435
[16] 田原，吴更石，梁德群．计算机人脸识别的研究现状与进展．无线电工程，1997，27(3)：5～14
[17] Cheng Y Q, Zhuang Y M, Wang H F, et al. An efficient recognition approach to human faces. ACTA Automatica Sinica, 1993, 19(1): 54～61
[18] Zhou Jiliu. A study on automatic recognition system for human profite. Journal of Sichuan University(Natural Science Edition), 1994, 31(1):100～104
[19] Samal A, lyengar P A. Automatic recognition and analysis for human faces and facial expressions: A survey. Pattern Recognition, 1992, 25(2): 65～77
[20] Lee S Y, Ham Y K, Park R H. Recognition of human front faces using knowledge-based feature extraction and neuro fuzzy algorithm. Pattern Recognition, 1996, 29(11): 1863～1876
[21] Huang C L, Chen C W. Human facial feature extraction for face interpretation and recognition. Pattern Recognition, 1992, 25(12): 1435～1444
[22] H. A. Rowley, S. Baluja, T. Kanade, Neural network-based face detection, IEEE Trans. PAMI, 1998, 20(1): 23～38
[23] K. K. Sung, T. Poggio, Example based learning for view-based human face detection, IEEE Trans. PAMI, 1998, 20(1): 39～50
[24] Peng Z, Too L, Xu G, et al. Detecting facial feature on images with multiple faces. In: Proc. conf. on Multimodal Interfaces, Beijing, China, 2000:191～198
[25] Tao L. M. Color perception for machine vision and human skin-color features: Ph. D thesis. Beijing: Tsinghua University. 2001. (in Chinese)
[26] Hong W, Ai H, Zhuang L, et al. Video based face verification.
In: Proc. of SPIE, 2001, 45(50):171～175
[27] Liang L, Ai H, He K, et al. Face detection based on the matching of multiple related templates, Chinese Journal of Software, 12(1): 94～102
[28] 边肇祺，张学工．模式识别．北京：清华大学出版社，2000：212～228
[29] Rowley, HA, S. Baluja, T. Neural Network-Based Face Detection, IEEE Transactions on PAMI, 1998, 20(1):23～38
[30] LIANG, Haizhou AI, Kezhong HE, et al. Multiple Related Template Matching Based Face Detection. Chinese Journal of Software, 2001, 12(1): 94～102.
[31] P. Viola & M. Jones, Rapid Object Detection using a Boosted Cascade of Simple Features, Proc. of IEEE Conf. on Computer Vision and Pattern Recognition, 2001, 12(1):511～518
[32] D. Graupe, Principles of Artificial Neural Networks, World Scientific Publishing Co. 1997, 15(3): 191～222
[33] R. Feraud, O. J. Bernier, J. E. Viallet, M. Collobert. A fast and accurate face detector based on neural networks, IEEE Trans. PAMI, 2001, 23(1):42～53
[34] L. Breiman. Bagging predictors. Machine Learning, 1996, 26(2): 123～140
[35] P. Buhlmann and B. Yu. Analyzing bagging. Ann. of Statist, 2002, 30(6):927～961
[36] S. X. Chen and P. Hall. Effects of bagging and bias correction on estimators defined by estimating equations. Statistica Sinica, 2003, 13(30): 97～109
[37] Patrik O. Hoyer. Non-negative Sparse Coding. Nin Proc. IEEE Workshop on Neural Networks for Signal Processing. 2002, 7(5): 557～565
[38] Qi X F, Palmieri F. Theoretical analysis of evolutionary algorithms with an infinite population size in continuous space, Part Ⅱ: Basic properties of selection and mutation. IEEE Trans on Neural Networks, 1994, 5(1): 120～129
[39] D. D. Lee, H. S. Seung. Algorithms for non-negative matrix
factorization. In Advances in Neural Information Processing Systems MIT Press, 2001, 13(7): 556～562
[40] 樊重俊，王浣尘．遗传算法的改进与应用．上海交通大学学报，1998，32(12)：127～131
[41] 夏利民，谷士文，宁国英．基于小波包分析的货车滚动轴承故障诊断．中国铁道科学，2002，23(6)：60～64
[42] 张俊，何昕，李介谷．基于面部几何特征点提取的人脸识别方法．红外与激光工程，1999，28(4)：40～44
[43] 范宏深，倪国强，申会堂．利用几何特性及神经网络进行人脸探测技术的研究．光学技术，2002，28(2)：105～108
[44] 梁路宏，艾海舟，何克忠．基于多模板匹配的单人脸检测．中国图像图形学报，1999，4(A)(10)：825～829
[45] 刘小军．人脸识别技术研究：[博士学位论文]．北京：中国科学院电子学研究所，2001，7
[46] Schapire, R. E., Freund, Y., Bartlett, P. et al. Boosting the margin: A new explanation for the effectiveness of voting methods. In Machine Learning: Proceedings of Fourteenth International Conference, 1997, 25(2): 322～330
[47] W. Zhao, R. Chellappa, A. Krishnaswamy. Discriminant analysis of principal components for face recognition. In International Conference on Automatic Face and Gesture Recognition 1998, 4(1): 336～341
[48] Alex Pentland, Baback Moghaddam, rhad Starner. View-Based and Modular Eigenspaces for Face Recognition. IEEE Conf. on Computer Vision and Pattern Recognition, MIT Media Laboratory Tech. Report 1994, 26(6): 245～249
[49] Breiman L. Bagging predictors. Machine Learning, 1996, 26(2): 123～140
[50] S. D. Bay. Nearest Neighbor Classification from Multiple Feature Subsets. Intelligent Data Analysis, 1999, 3(2): 191～209
[51] D. Lee, H. Seung. Learning the parts of objects by non-negative matrix factorization. Natur, 1999, 401(3): 788～791
[52] Thierry Denoeux. A k-Nearest Neighbor Classification Rule
Based on Dempster2Shafer Theory[J]. IEEE Trans. on Systems Man and Cybernetics, 1995, 25(5): 804～813
[53] Aleix M, Mart nez, and Avinash C. Kak. PCA versus LDA. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001, 23(2): 228～233
[54] Bauer, E, Kohavi R. An Empirical Comparison of Voting Classification Algorithms: Bagging, Boosting, and Variants. Machine Learning, 1999, 36(3): 105～139
[55] R. Feraud, O. J. Bernier. A fast and accurate face detection based on neural network. IEEE PAMI, 2001, 23(1): 42～53
[56] G. Donato, M. S. Bartlett, J. C. Hager, P. Ekman, et al. Sejnowski, "Classifying Facial Actions," IEEE Transactions on Pattern Analysis and Machine Intelligence, 1999, 21(2): 974-989