基于肤色模型和Adaboost算法的人脸检测系统
|
摘要
人脸检测是指将人脸从图像或视频中检测出来并提取面部特征的过程,它是人脸识别、特征定位的首要环节。同时在监测跟踪、出入检查、医疗领域也有重要应用价值。
人脸检测受背景、光线及头部姿势等因素的影响而变的复杂。本文在对人脸检测技术进行深入学习和研究的基础上,主要对图像去噪声,检测特征选取和检测精度提高三方面问题展开研究。
针对图像噪声问题,采用“参考白”算法、均值滤波算法对图像进行光线补偿和平滑处理,消除了图像中的光线和噪点干扰。采用中值法分离出视频背景,采用帧间差分算法提取出包含人脸的运动前景,去除了冗余视频背景,提高了检测速度。选取肤色作为人脸检测特征,提出了样本提取与Fisher准则结合检测肤色的方法,利用非线性变换改进了肤色在RGB与YC_bC_r色彩空间的聚类性,求得肤色在RGB与YC_bC_r色彩空间的样本高斯分布,计算出肤色似然图。利用Fisher准则分析肤色似然图,求出分割阈值,将图像二值化,划分二值化图连通区域得到肤色窗口。针对肤色检测精度不高问题,应用改进的Adaboost算法训练出由三个强分类器组成级联分类器,精检测肤色窗口。各强分类器由haar弱分类器加权构成。最后提出用图像横向积分与连通域结合法提取出眼睛特征,利用对像素R色彩通道分析提出嘴唇特征。
结合改进的肤色提取与Adaboost算法,可以有效滤除干扰,并提高了检测速度,降低了误检率,保证了较高的人脸检测率。研究工作有一定理论意义和应用价值。
Face Detection is a process that detecting the human faces from images or videos and distill the characters. It is a key point of Face recognition and Facial feature extract. At the same time, It has important applicable value in some places such like track, pass check and medical treatment.
Face Detection is a complex problem because of it influenced by background, light and head's posture. Based on the study on Face Detection technology, Noise Filter, Detection Character and Precision Improve are the main research directions in this paper.
Compensate the picture's light and then filter the interference color by using reference white arithmetic and average evaluation arithmetic for filter the noise. Median arithmetic is used to separate the background, Median difference arithmetic is used to separate moving information, filter the background and expedite the detection. Choose the complexion as detection character. bring forward a arithmetic that synthesize Stylebook Distill and Fisher judgment. Improve the complexion clustering in RGB and YC_bC_r color space and gained the complexion distributing. Account out the resemble image. Analyze the resemble image by fisher judgment and gained the binary image. Search the connect areas and get the complexion windows. Complexion Detection's not high precision, training a series sort implement include three strong sort implements to detect the complexion windows subtly. Each strong sort implement is buildup with lots of weak sort implements. Finally detect the eyes character by the complexion windows' landscape orientation integral. Detect the lips character by analyzing the R channels signal.
By using complexion and Adaboost arithmetic can eliminate the noise signal effectively. It can both pledge the high detection rate and the low error judge rate and improve the system's operation speed. The research has some theory and application value.
人脸检测受背景、光线及头部姿势等因素的影响而变的复杂。本文在对人脸检测技术进行深入学习和研究的基础上,主要对图像去噪声,检测特征选取和检测精度提高三方面问题展开研究。
针对图像噪声问题,采用“参考白”算法、均值滤波算法对图像进行光线补偿和平滑处理,消除了图像中的光线和噪点干扰。采用中值法分离出视频背景,采用帧间差分算法提取出包含人脸的运动前景,去除了冗余视频背景,提高了检测速度。选取肤色作为人脸检测特征,提出了样本提取与Fisher准则结合检测肤色的方法,利用非线性变换改进了肤色在RGB与YC_bC_r色彩空间的聚类性,求得肤色在RGB与YC_bC_r色彩空间的样本高斯分布,计算出肤色似然图。利用Fisher准则分析肤色似然图,求出分割阈值,将图像二值化,划分二值化图连通区域得到肤色窗口。针对肤色检测精度不高问题,应用改进的Adaboost算法训练出由三个强分类器组成级联分类器,精检测肤色窗口。各强分类器由haar弱分类器加权构成。最后提出用图像横向积分与连通域结合法提取出眼睛特征,利用对像素R色彩通道分析提出嘴唇特征。
结合改进的肤色提取与Adaboost算法,可以有效滤除干扰,并提高了检测速度,降低了误检率,保证了较高的人脸检测率。研究工作有一定理论意义和应用价值。
Face Detection is a process that detecting the human faces from images or videos and distill the characters. It is a key point of Face recognition and Facial feature extract. At the same time, It has important applicable value in some places such like track, pass check and medical treatment.
Face Detection is a complex problem because of it influenced by background, light and head's posture. Based on the study on Face Detection technology, Noise Filter, Detection Character and Precision Improve are the main research directions in this paper.
Compensate the picture's light and then filter the interference color by using reference white arithmetic and average evaluation arithmetic for filter the noise. Median arithmetic is used to separate the background, Median difference arithmetic is used to separate moving information, filter the background and expedite the detection. Choose the complexion as detection character. bring forward a arithmetic that synthesize Stylebook Distill and Fisher judgment. Improve the complexion clustering in RGB and YC_bC_r color space and gained the complexion distributing. Account out the resemble image. Analyze the resemble image by fisher judgment and gained the binary image. Search the connect areas and get the complexion windows. Complexion Detection's not high precision, training a series sort implement include three strong sort implements to detect the complexion windows subtly. Each strong sort implement is buildup with lots of weak sort implements. Finally detect the eyes character by the complexion windows' landscape orientation integral. Detect the lips character by analyzing the R channels signal.
By using complexion and Adaboost arithmetic can eliminate the noise signal effectively. It can both pledge the high detection rate and the low error judge rate and improve the system's operation speed. The research has some theory and application value.
引文
[1]Yullle A L.Deformable Templates for Face Recognition[J].Journal of Cognitive Neuroscience,1991,3(1):59-70.
[2]梁路宏,艾海舟,何克忠等.基于多模板匹配的单人脸检测[J].中国图象图形学报,1999,4A(10):823-830.
[3]Turk M,Pentland A.Eigen face for Recognition[J].Cognitive Neuroscience,1991,23(3):71-86.
[4]Moghaddma B,Pentland A.Probabilistic Visual Learning for object Representation.IEEE Trans on PAMI,1997,19(7):697-709.
[5]张学.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):31-41.
[6]陈刚,戚飞虎.实用人脸识别系统的本征脸法实现[J].计算机研究与发展,2001,38(2):170-175
[7]Wu H Y,Chen Q,Yeshiva M.Face detection from color images using a fuzzy Pattern matching method.IEEE Trans Pattern Analysis and Machine Intelligence,1999,21(6):557-563.
[8]Rowley H A,Beluga S,Kanade T.Neural Network Based Face Detention[J].IEEE Irans PMAI,1998,20(i):23-38.
[9]Kouzani AZ,He F,Samut K.Commonsense knowledge based face detection.Pore Conference on Intelligent Engineering systems,Budapest,Hungary,1997,215-220.
[10]荆晓远,金忠,杨静宇.基于带通滤波和遗传算法的人脸图像预处理及识别[J].中国图像图形学报,1998,3(10):840-844.
[11]Swetes D L,Weng J.Using Discriminate Eigen features for Image Retrieval[J].IEEE Trans PMAI,1996,18(8):831-836.
[12]Luls T,Josep V.Automatic Face Recognition for Video Indexing Applications[J].Pattern Recognition,2002,35(3):615-625.
[13]AlbertP,Jordi Vitriol.Topological Principal Component Analysis for Face Encoding and Recognition[J].Pattern Recognition Letters,2001,22(6.7):769-776.
[14]Paul Viola,Michael Jones.Rapid Object Detection using a Boosted Cascade of Simple Features.IEEE Computer Vision and Pattern Recognition(CVPR01),VOL1,2001.
[15]Yang G Z,Huang T S.Human face detection in a complex background.Pattern Recognition,1994,27(1):53-61.
[16]卢春雨,张长水,闻芳.基于区域特征的快速人脸检测法[J].清华大学学报(自然科学版),1999,39(1):101-105.
[17]Wangzeng Kong,Shanan Zhu.A New Method of Single Color Face Detection Based on Skin Model and Gaussian Distribution,the 6~(th) world Congress on Intelligent Control and Automation,June21-23,2006.
[18]Schneider man H,Kanade T.Probabilistic modeling of local appearance and spatial elation ships for object recognition.In:Proc IEEE Conference on Computer Vision and Pattern Recognition,Santa Barbara,Califomia,1998:45-51.
[19]王耀明,图像的矩函数原理、算法及应用[M].上海:华东理工大学出版社,2002.
[20]Froba B,A.Emst.Fast Frontal View Face Detection Using a Multi-Path Decision Tree.In:Proc Analysis and Video based Biometric Person Authentication(AVBPA2003),2003:921-928.
[21]A ogihara,A Shintani,S Takematsu.Speech Recognition based on the fusion of Visual and Auditory information using full-Frame color image.IEEE Yrans Fundamentals,1996:1836-1840.
[22]Boxer B E,Guyon I M.Vapnik VNA training algorithm for optimal margin classifier.In:Proc of the 5~(th) ACM Workshop on Computational Learning Theory,Pittsburgh,1992:144-152.
[23]李弼程,彭天强,彭波.智能图像处理技术[M].北京:北京电子工业出版社.2004
[24]Terrillon J C,Shirazi M N.Comparative performance of different skin Chrominance models and chrominance spaces for the automate deletion of human faces In Color images.In:Proc.Conf.on Automatic Face and Gesture Recognition.Granola,France,2000:54-61.
[25]M.Hu,S.W6rrall,A.H.Sadka.Face Feature Detection and Model Design For 2.D Scalable Model.Based Video Coding,VIE,Guildford.London,UK July 2003.
[26]Valiant L.A theory of the learnable Artificial Intelligence and Language Processing.1982(27):1134-1142.
[27]Vladimir N Vapnik.Statistical Learning Theory[M].Wiely,New York,1998.
[28]Kearns M.The Computational Complexity of Machine Learning.Cambridge:MIT Press.1990.
[29]Sehapire R.E.The Strength of weak Learn ability.Machine Learning,1990.5(2):197-227,
[30]Freund,Y,R.E.SchaPire.Experiments with a new boosting algorithm,Machine Learning Proceedings of the Thirteenth International Conference.Bari,Italy,1996:148-156.
[31]Freund,Y,R.E.SchaPire.Game theory On-line Prediction and boosting.Proceedings of the Ninth Annual Conference on Computational Learning Theory.Dissension de1 Grade,Italy,1996:325-332.
[32]C Pageorgiou,M.Oren.A general framework for object detection.In:International Conference on Computer Vision,1998.
[33]R.E.SchaPire.The Boosting approach to machine learning,an overview MSRI workshop on Nonlinear Estimation and Classification,2002.
[34]Ferand R,Bernier O.A fast and accurate face detector based on neural networks.IEEE Transactions on Pattern Analysis and Machine Intelligence,2001,23(1):42-53.
[35]Ii S.Z,Zhu L,Zhang,Z.Q.Statistical learning of multi-view face detection.In:Proceedings of the 7~(th) European Conference on Computer Vision Copenhagen,2002.
[36]Lienhart R,Maydt J.An extended set of Haar-like features for Rapid object detection.IEEE International conference on Image processing,New York,USA,2002(1):903.
[37]Pentland A,Moghaddam B.View based and modular Eigen Spaces for face recognition.In:Proceedings of the IEEE Conference Computer Vision and Pattern Recognitions,Seattle,WA,1994:21-23.
[38]Y Freund.Boosting a weak learning algorithm by majority,Information and Computer.121,No.2 1995:256-285.
[39]Crow F C.Summed-area tables for texture mapping.SIGGRAPH,1984:207-212.,
[40]Sochman J,Matas J.Adaboost with totally corrective updates for fast face detection.AFGR04,2004:445-450.
[41]Yow K.C,Cibola R.Feature based human face detections.Image and Vision Computing,1997,15(9):713-735.
[42]Burgess C.J.C.A tutorial on support vector machines for Pattern recognition.Data Mining and Knowledge Discovery,1998,2(2):121-167.
[43]Y Gong,M Sakauchi.Detection of regions matching specified chromatic features[J].Computer Vision and Image Understanding,1995,61(2):263-269.
[44]Scott Still man,Q Alex Zhao,Xin Zhong.The Smart Room Final Report.Fall 1997.
[45]章毓晋.图像分割[M].北京:科学出版社,2001.
[46]Peng H,Zhang D.Dual Eigen space Method for Human Face Recognition[J].Electronics Letters,1997,33(4):283-284.
[47]边肇棋,张学工等.模式识别(第二版)[M].北京:清华大学出版社,2001:87-90.
[48]Moal P,JnoesM.Rapid object detection using a boosted cascade of simple features.In:Proceedings of IEEE Conference Computer Vision and Pattern Recognition,Kauai,Hawaii,USA,2001:511-518.
[49]洪子泉,杨静宇.统计模式识别中的特征提取[J].数据采集与处理,1991,6(2):38-43.
[50]张宏林,蔡锐.VisualC++图像模式识别技术及工程实践[M].北京:人民邮电出版社,2001.
[2]梁路宏,艾海舟,何克忠等.基于多模板匹配的单人脸检测[J].中国图象图形学报,1999,4A(10):823-830.
[3]Turk M,Pentland A.Eigen face for Recognition[J].Cognitive Neuroscience,1991,23(3):71-86.
[4]Moghaddma B,Pentland A.Probabilistic Visual Learning for object Representation.IEEE Trans on PAMI,1997,19(7):697-709.
[5]张学.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):31-41.
[6]陈刚,戚飞虎.实用人脸识别系统的本征脸法实现[J].计算机研究与发展,2001,38(2):170-175
[7]Wu H Y,Chen Q,Yeshiva M.Face detection from color images using a fuzzy Pattern matching method.IEEE Trans Pattern Analysis and Machine Intelligence,1999,21(6):557-563.
[8]Rowley H A,Beluga S,Kanade T.Neural Network Based Face Detention[J].IEEE Irans PMAI,1998,20(i):23-38.
[9]Kouzani AZ,He F,Samut K.Commonsense knowledge based face detection.Pore Conference on Intelligent Engineering systems,Budapest,Hungary,1997,215-220.
[10]荆晓远,金忠,杨静宇.基于带通滤波和遗传算法的人脸图像预处理及识别[J].中国图像图形学报,1998,3(10):840-844.
[11]Swetes D L,Weng J.Using Discriminate Eigen features for Image Retrieval[J].IEEE Trans PMAI,1996,18(8):831-836.
[12]Luls T,Josep V.Automatic Face Recognition for Video Indexing Applications[J].Pattern Recognition,2002,35(3):615-625.
[13]AlbertP,Jordi Vitriol.Topological Principal Component Analysis for Face Encoding and Recognition[J].Pattern Recognition Letters,2001,22(6.7):769-776.
[14]Paul Viola,Michael Jones.Rapid Object Detection using a Boosted Cascade of Simple Features.IEEE Computer Vision and Pattern Recognition(CVPR01),VOL1,2001.
[15]Yang G Z,Huang T S.Human face detection in a complex background.Pattern Recognition,1994,27(1):53-61.
[16]卢春雨,张长水,闻芳.基于区域特征的快速人脸检测法[J].清华大学学报(自然科学版),1999,39(1):101-105.
[17]Wangzeng Kong,Shanan Zhu.A New Method of Single Color Face Detection Based on Skin Model and Gaussian Distribution,the 6~(th) world Congress on Intelligent Control and Automation,June21-23,2006.
[18]Schneider man H,Kanade T.Probabilistic modeling of local appearance and spatial elation ships for object recognition.In:Proc IEEE Conference on Computer Vision and Pattern Recognition,Santa Barbara,Califomia,1998:45-51.
[19]王耀明,图像的矩函数原理、算法及应用[M].上海:华东理工大学出版社,2002.
[20]Froba B,A.Emst.Fast Frontal View Face Detection Using a Multi-Path Decision Tree.In:Proc Analysis and Video based Biometric Person Authentication(AVBPA2003),2003:921-928.
[21]A ogihara,A Shintani,S Takematsu.Speech Recognition based on the fusion of Visual and Auditory information using full-Frame color image.IEEE Yrans Fundamentals,1996:1836-1840.
[22]Boxer B E,Guyon I M.Vapnik VNA training algorithm for optimal margin classifier.In:Proc of the 5~(th) ACM Workshop on Computational Learning Theory,Pittsburgh,1992:144-152.
[23]李弼程,彭天强,彭波.智能图像处理技术[M].北京:北京电子工业出版社.2004
[24]Terrillon J C,Shirazi M N.Comparative performance of different skin Chrominance models and chrominance spaces for the automate deletion of human faces In Color images.In:Proc.Conf.on Automatic Face and Gesture Recognition.Granola,France,2000:54-61.
[25]M.Hu,S.W6rrall,A.H.Sadka.Face Feature Detection and Model Design For 2.D Scalable Model.Based Video Coding,VIE,Guildford.London,UK July 2003.
[26]Valiant L.A theory of the learnable Artificial Intelligence and Language Processing.1982(27):1134-1142.
[27]Vladimir N Vapnik.Statistical Learning Theory[M].Wiely,New York,1998.
[28]Kearns M.The Computational Complexity of Machine Learning.Cambridge:MIT Press.1990.
[29]Sehapire R.E.The Strength of weak Learn ability.Machine Learning,1990.5(2):197-227,
[30]Freund,Y,R.E.SchaPire.Experiments with a new boosting algorithm,Machine Learning Proceedings of the Thirteenth International Conference.Bari,Italy,1996:148-156.
[31]Freund,Y,R.E.SchaPire.Game theory On-line Prediction and boosting.Proceedings of the Ninth Annual Conference on Computational Learning Theory.Dissension de1 Grade,Italy,1996:325-332.
[32]C Pageorgiou,M.Oren.A general framework for object detection.In:International Conference on Computer Vision,1998.
[33]R.E.SchaPire.The Boosting approach to machine learning,an overview MSRI workshop on Nonlinear Estimation and Classification,2002.
[34]Ferand R,Bernier O.A fast and accurate face detector based on neural networks.IEEE Transactions on Pattern Analysis and Machine Intelligence,2001,23(1):42-53.
[35]Ii S.Z,Zhu L,Zhang,Z.Q.Statistical learning of multi-view face detection.In:Proceedings of the 7~(th) European Conference on Computer Vision Copenhagen,2002.
[36]Lienhart R,Maydt J.An extended set of Haar-like features for Rapid object detection.IEEE International conference on Image processing,New York,USA,2002(1):903.
[37]Pentland A,Moghaddam B.View based and modular Eigen Spaces for face recognition.In:Proceedings of the IEEE Conference Computer Vision and Pattern Recognitions,Seattle,WA,1994:21-23.
[38]Y Freund.Boosting a weak learning algorithm by majority,Information and Computer.121,No.2 1995:256-285.
[39]Crow F C.Summed-area tables for texture mapping.SIGGRAPH,1984:207-212.,
[40]Sochman J,Matas J.Adaboost with totally corrective updates for fast face detection.AFGR04,2004:445-450.
[41]Yow K.C,Cibola R.Feature based human face detections.Image and Vision Computing,1997,15(9):713-735.
[42]Burgess C.J.C.A tutorial on support vector machines for Pattern recognition.Data Mining and Knowledge Discovery,1998,2(2):121-167.
[43]Y Gong,M Sakauchi.Detection of regions matching specified chromatic features[J].Computer Vision and Image Understanding,1995,61(2):263-269.
[44]Scott Still man,Q Alex Zhao,Xin Zhong.The Smart Room Final Report.Fall 1997.
[45]章毓晋.图像分割[M].北京:科学出版社,2001.
[46]Peng H,Zhang D.Dual Eigen space Method for Human Face Recognition[J].Electronics Letters,1997,33(4):283-284.
[47]边肇棋,张学工等.模式识别(第二版)[M].北京:清华大学出版社,2001:87-90.
[48]Moal P,JnoesM.Rapid object detection using a boosted cascade of simple features.In:Proceedings of IEEE Conference Computer Vision and Pattern Recognition,Kauai,Hawaii,USA,2001:511-518.
[49]洪子泉,杨静宇.统计模式识别中的特征提取[J].数据采集与处理,1991,6(2):38-43.
[50]张宏林,蔡锐.VisualC++图像模式识别技术及工程实践[M].北京:人民邮电出版社,2001.