基于肤色和人脸特征的人脸检测和人眼定位方法研究
|
摘要
基于生物特征的身份认证技术近年来发展迅速。相对于其它生物认证技术,人脸识别具有更方便、直接、友好的优点,成为当前热门研究领域之一。人脸检测是人脸识别过程中的重要前期工作,其主要任务是将人脸部分从背景图像中分离出来。特别是彩色图像中的人脸检测问题成为近几年模式识别、机器视觉研究领域的热点问题,其被认为是图像分析和理解技术典型的应用之一。它融合了模式识别、机器视觉、人工智能以及图像处理等许多先进技术。人脸检测系统可以被应用到数码产品、安全监控系统、人机交互等系统中,具有十分广泛的应用前景和商业价值。
在彩色图像中,肤色是人脸的重要信息。它不依赖于面部的细节特征,具有相对的稳定性。在色彩空间中,人脸的肤色分布表现出良好的聚类特性。根据肤色特征可以直接去除大量背景因素迅速得到目标区域。
本文描述了一种能够对不同背景下彩色图像快速定位人眼的框架。首先,对图像进行光照补偿,然后,根据YCbCr颜色空间肤色聚类的特点和椭圆模型分割图像,检测人脸区域,并利用基于“像素密度”的滤波方法去除大面积的噪声。最后,利用人脸的结构特征和对称信息准确定位人眼。实验结果表明,该方法能够准确定位人眼,并且对于具有不同背景,带眼镜的,表情变化的图像都能有很好的鲁棒性。
The technology of identity authentication that bases on biological characters is developing quickly. Compared with other biometrics technology, face recognition is more convenient ,direct and friendly. So, it becomes to be one of the hottest research domains. Face detection is an important pre-processing stage, and the main function is to separate the face from the background. Face detection technology, as one of the typical applications of image analysis and understanding, is an important research subject among computer vision and pattern recognition areas. Its research involves in many advanced technologies, such as pattern recognition , computer vision, artificial intelligence,image processing. Many real applications, for example, digital products, safety surveillance systems, human-computer interaction, adopt face detection technology; therefore this technology possesses great commercial and usage value.
In color images, the skin-color is important information of face. It is stable, and independent of detail's characteristics. The skin-color has a good character of clustering in color space. According to this character, much background factor can be discarded directly.
This paper proposes a method that is capable of segmenting color archives images and locating eyes rapidly to the different background. Firstly, execute illumination compensation for images. Then, segment the image according to the skin-color’s clustering characteristic in the YCbCr color space and elliptical model, and detect the face region, and use the filtering method based on 'the pixel density' to remove the noise of big area. Lastly, utilize the structural characteristic and symmetrical information of human face to locate the eyes accurately. Experimental results prove that this method could locate eyes exactly, and it is robust for the images with different background, bespectacled, as well as images existing expression changes.
在彩色图像中,肤色是人脸的重要信息。它不依赖于面部的细节特征,具有相对的稳定性。在色彩空间中,人脸的肤色分布表现出良好的聚类特性。根据肤色特征可以直接去除大量背景因素迅速得到目标区域。
本文描述了一种能够对不同背景下彩色图像快速定位人眼的框架。首先,对图像进行光照补偿,然后,根据YCbCr颜色空间肤色聚类的特点和椭圆模型分割图像,检测人脸区域,并利用基于“像素密度”的滤波方法去除大面积的噪声。最后,利用人脸的结构特征和对称信息准确定位人眼。实验结果表明,该方法能够准确定位人眼,并且对于具有不同背景,带眼镜的,表情变化的图像都能有很好的鲁棒性。
The technology of identity authentication that bases on biological characters is developing quickly. Compared with other biometrics technology, face recognition is more convenient ,direct and friendly. So, it becomes to be one of the hottest research domains. Face detection is an important pre-processing stage, and the main function is to separate the face from the background. Face detection technology, as one of the typical applications of image analysis and understanding, is an important research subject among computer vision and pattern recognition areas. Its research involves in many advanced technologies, such as pattern recognition , computer vision, artificial intelligence,image processing. Many real applications, for example, digital products, safety surveillance systems, human-computer interaction, adopt face detection technology; therefore this technology possesses great commercial and usage value.
In color images, the skin-color is important information of face. It is stable, and independent of detail's characteristics. The skin-color has a good character of clustering in color space. According to this character, much background factor can be discarded directly.
This paper proposes a method that is capable of segmenting color archives images and locating eyes rapidly to the different background. Firstly, execute illumination compensation for images. Then, segment the image according to the skin-color’s clustering characteristic in the YCbCr color space and elliptical model, and detect the face region, and use the filtering method based on 'the pixel density' to remove the noise of big area. Lastly, utilize the structural characteristic and symmetrical information of human face to locate the eyes accurately. Experimental results prove that this method could locate eyes exactly, and it is robust for the images with different background, bespectacled, as well as images existing expression changes.
引文
[1]Rowley H A,Baluja S,Kanade T. Rotation invariant neural network-based face detection ,CMU-CS-97-201,1997.
[2]Rowley H A,Baluja S,Kanade T.Rotation Invariant Neural Network Based Face Detection[C].In:Proceeding of IEEE Computer on CVPR’98.california.1998.
[3]Rowley H A, Baluja S,Kanade T.Neural network-based face detection. IEEE Trans.Pattern Analysis and Machine Intelligence,1998,20(l):23-38
[4]Rowley H A.Neural network-based face deteetion. Ph.D.thesis,CMU-CS-99-117,1999.
[5]R.Brunelli,T.Poggio,FaceRecognition Through Geometrical Features[R],Proceedings of ECCV,PP.792-800,1992
[6]N.Intrator,D.Reisfeld,Y.Yeshurun,Extraction of Facial Features for Recognition Using Neural Networks[R],Proceedings of International Workshop on Automatic Face and Gesture Recognition,pp.260-265,1995
[7]H.Zabrodsky et al.IEEE Trans.PAMI,17:1154-1166,1995
[8]艾海舟,梁路宏,徐光佑等.基于肤色和模板的人脸检测[J].软件学报,2001,12.12:1784-1792.
[9]梁路宏,艾海舟,肖习攀等.基于模板匹配与支持矢量机的人脸检测[J].计算机学报,2002,25.1:22-29.
[10]刘翼光,沈理.利用Hausdroff距离人脸图像定位算法[J].计算机研究与发展,2001,38.4:475-481.
[11]Yuille A,Hallinan P,Cohen D.Feature Extraction from Faces Using Deformable Templates[J].International Journal of Computer Vision.1992,8.2:99-111.
[12]马宇飞,白雪生,徐光佑等.新闻视频中口播帧检测方法的研究[J].软件学报,2001,12.3:377-381.
[13]山世光,高文,陈熙林.基于纹理分布和变形模板得面部特征提取[J].软件学报,2001,12.4:570-577.
[14]Valentin D ,Abdi H, O’Toole A J,et al .Connectionist models of face processing :a survey ,Pattern Recognition,27(9):1209-1230,1994
[15]Burel G , Carel D ,Detection and localization of faces on digital images ,Pattern Recognition Letters,15(10):963-967,1994
[16]Karlekar J ,Marsh R , A hierarchical neural network for human face detection, Pattern Recognition, 29(5):781-787,1996
[17]SungKK,Poggio T,Example-based learning for view-based human face detection, I- EEE Transactions on Pattern Analysis and Machine Intelligence,20(1):39-50,1998
[18] Rowley H,Baluja S,Kanade T,Rotation invariant neural network-based face detection , in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition , Santa Barbara , California, USA,963-969,1998
[19] Fast Multi-view Face Detection
[20]C.J.CBurges.ATutorial on Support Vector Machines for Pattern Recognition[J].Data Mining and Knowledge Discovery,2(2):1-47,1998
[21]H Schneiderman,T Kanade.Probabilistic Modeling of Local Appearance and Spatial Relationships for Object Recognition[C].In:Proc IEEE Conf Computer Vision and Pattern Recognition,1998:45-51
[22]Nefian A V.A Hidden Markov Model-Based Approach for Face Detection and Recognition[J].school of Electrical and Computer Engineering,Georigia Institute of Technology,1998
[23]A.V.Nefian,M.H.HayesⅢ.An embedded HMM based approach for face detection and recognition[C],IEEE Inter.Conf.on Acoustics,Speech and Signal Processing, Phoen- ix AZ.1999
[24]B.D. Zarit, B. J. Super, F.K.H. Quek. Comparison of Five Color Models in Skin Pixel Classification [A]. International Workshop on Recognition, Analysis and Tracking of Faces and Gestures in Real-Time Systems [C], 1999: 58-63.
[25] L.Jordao, M.Perrone, J.Costeira, J.Santos-Victor.Active Face and Feature Tracking [A]. Proceedings of the 10th International Conference on Image Analysis and Processing [C], 1999: 572-577.
[26] L.Sigal, S.Sclaroff,V Athitsos.Estimation and Prediction of Evolving Color Distributions for Skin Segmentation under Varying Illumination [A]. Proceeding of IEEE Conference on Computer Vision and Pattern Recognition [C], 2000, 2:152-159.
[27] S.Mckenna, S.Gong,Y Raja.Modeling Facial Colour and Identity with Gaussian Mixtures[J].Pattern Recognition,1999,31(12):1883-1892.
[28]Visual C++数字图像模式识别技术及工程实践[M].北京:人民邮电出版社,2003年2月版.
[29]J.Kovae. P. Peer, F. Solina. Human Skin Colour Clustering for Face Detection[A].E- UROCON 2003, Computer as a Tool[C],2003:144-148.
[30]L.Jordao, M.Perrone, J.Costeira, J.Santos-Victor. Active Face and Feature Tracking [A].Proceedings of the 10th International Conference on Image Analysis and Processing [C], 1999: 572-577.
[31]J.Ahlberg.A System for Face Localization and Facial Feature Extraction [R]Techni- cal Report, LiTH-ISY R-2172, Linkoping University, 1999.
[32]R-L.Hsu,A-M.Mohamed,A.K.,Jain.Face Deteetion in Color Images[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(5):696-706.
[33]J.Ahlberg.A System for Face Localization and Facial Feature Extraction [R].Technical Report,LiTH-ISY-R-2172,Linkoping Univerity,1999.
[34] M.Yang,N.Ahuja. Gaussian mixture model for human skin color and its application in image and video databases[A].Proceedings of the SPIE: Conference on Storage and Retrieval for Image and Video Databases[C],1999,3656:458-466.
[35]M.J.JONES , J.M.REHG. Statistical color models with application to skin detection[A]. Proeeedings of IEEE Conference on Computer Vision and Pattern Recognition[C],1999:274-280.
[36] M.Yang,N.Ahuja. Gaussian mixture model for human skin color and its application in image and video databases[A].Proceedings of the SPIE: Conference on Storage and Retrieval for Image and Video Databases[C],1999,3656:458-466.
[37]J-C.Terrillon,M.N.Shirazi,H.Fukamaehi,S.Akamatsu. Comparative Performance of Differeni Skin Chrominance Models and Chrominance Spaces for the AutomaticDetection of Human Faces in Color Images[A].Proceeding of the International Conference on Face and Gesture Recognition[C」,2000,54一61.
[38]N.Oliver,A.Pentland,F.Berard.Lafter:Lips and Face Real Time Tracker with Facial Expression Recognition[A].Proeeedings of Computer Vision and Pattern Recognition[c] 1997:123-129.
[39]田捷,杨鑫.生物特征识别技术理论与应用[M].2005,北京,电子工业出版社.
[40] Rafael C. Gonzalez Richad E. Woods.数字图像处理(第一版)[M].阮秋琦,阮宇智.北京.电于工业出版社,2003.
[41]Hsu Rdirr-Lien,Mohamed,Jain Anil K. Face detection in color images[J].IEEE Trans on Pattern Analysis and Machine Intelligence,2002,24(5):696-706.
[42]《A System for human face and facial feature location》.
[43]Visual C++数字图像模式识别技术及工程实践[M].北京:人民邮电出版社,2003年2月版.
[44]汪更生,林家骏,刘云翔,基于色度信息及投影法的彩色图像中人脸区域分割方法,长春工程学院学报(自然科学版),2004,5(3):51一54.
[45]E. Osuna. Support Vector Machines: Training and Applications.PhD thesis, MIT, EE/CS Dept., Cambridge, MA,1998.
[46]Visual C++数字图像模式识别技术及工程实践[M].北京:人民邮电出版社,2003年2月版.
[47]张志刚,周明全,耿国华.人脸关键特征点自动标定研究[J].计算机工程与应用.2007,43(21):197-217.
[48]田捷,杨鑫.生物特征识别技术理论与应用[M].2005,北京,电子工业出版社.
[49]朱夏君。人脸识别的研究及眼睛定位算法〔C],硕士学位论文,浙江工业大学,2004
[50]段立娟,崔国勤,高文等,多层次特定图像过滤方法[J].计算机辅助设计与图形学报,2002,14(5):404409.
[51]衣晓飞,陈福接。单人图像中人眼的快速定位[[J],计算机工程与科学,2001 23 (3)
[52]石东海。人脸检测和定位方法研究[C],硕士学位论文,国防科学技术大学,2002 1
[53]冈萨雷斯。数字图像处理,第二版。电子工业出版社
[54]邓加琼.复杂背景下的彩色人脸检测算法与实现「D]。成都:电子科技大学,2005. 19-21
[55]王良民,张建明,詹永照,宋顺林。人脸检测研究现状和发展。江苏大学学报(自然科学版)第24卷第3期2003年5月
[56]唐伟,陈兆乾,吴建鑫,周志华。静态图像中的人脸检测方法综述。计算机科学,2002年20期
[2]Rowley H A,Baluja S,Kanade T.Rotation Invariant Neural Network Based Face Detection[C].In:Proceeding of IEEE Computer on CVPR’98.california.1998.
[3]Rowley H A, Baluja S,Kanade T.Neural network-based face detection. IEEE Trans.Pattern Analysis and Machine Intelligence,1998,20(l):23-38
[4]Rowley H A.Neural network-based face deteetion. Ph.D.thesis,CMU-CS-99-117,1999.
[5]R.Brunelli,T.Poggio,FaceRecognition Through Geometrical Features[R],Proceedings of ECCV,PP.792-800,1992
[6]N.Intrator,D.Reisfeld,Y.Yeshurun,Extraction of Facial Features for Recognition Using Neural Networks[R],Proceedings of International Workshop on Automatic Face and Gesture Recognition,pp.260-265,1995
[7]H.Zabrodsky et al.IEEE Trans.PAMI,17:1154-1166,1995
[8]艾海舟,梁路宏,徐光佑等.基于肤色和模板的人脸检测[J].软件学报,2001,12.12:1784-1792.
[9]梁路宏,艾海舟,肖习攀等.基于模板匹配与支持矢量机的人脸检测[J].计算机学报,2002,25.1:22-29.
[10]刘翼光,沈理.利用Hausdroff距离人脸图像定位算法[J].计算机研究与发展,2001,38.4:475-481.
[11]Yuille A,Hallinan P,Cohen D.Feature Extraction from Faces Using Deformable Templates[J].International Journal of Computer Vision.1992,8.2:99-111.
[12]马宇飞,白雪生,徐光佑等.新闻视频中口播帧检测方法的研究[J].软件学报,2001,12.3:377-381.
[13]山世光,高文,陈熙林.基于纹理分布和变形模板得面部特征提取[J].软件学报,2001,12.4:570-577.
[14]Valentin D ,Abdi H, O’Toole A J,et al .Connectionist models of face processing :a survey ,Pattern Recognition,27(9):1209-1230,1994
[15]Burel G , Carel D ,Detection and localization of faces on digital images ,Pattern Recognition Letters,15(10):963-967,1994
[16]Karlekar J ,Marsh R , A hierarchical neural network for human face detection, Pattern Recognition, 29(5):781-787,1996
[17]SungKK,Poggio T,Example-based learning for view-based human face detection, I- EEE Transactions on Pattern Analysis and Machine Intelligence,20(1):39-50,1998
[18] Rowley H,Baluja S,Kanade T,Rotation invariant neural network-based face detection , in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition , Santa Barbara , California, USA,963-969,1998
[19] Fast Multi-view Face Detection
[20]C.J.CBurges.ATutorial on Support Vector Machines for Pattern Recognition[J].Data Mining and Knowledge Discovery,2(2):1-47,1998
[21]H Schneiderman,T Kanade.Probabilistic Modeling of Local Appearance and Spatial Relationships for Object Recognition[C].In:Proc IEEE Conf Computer Vision and Pattern Recognition,1998:45-51
[22]Nefian A V.A Hidden Markov Model-Based Approach for Face Detection and Recognition[J].school of Electrical and Computer Engineering,Georigia Institute of Technology,1998
[23]A.V.Nefian,M.H.HayesⅢ.An embedded HMM based approach for face detection and recognition[C],IEEE Inter.Conf.on Acoustics,Speech and Signal Processing, Phoen- ix AZ.1999
[24]B.D. Zarit, B. J. Super, F.K.H. Quek. Comparison of Five Color Models in Skin Pixel Classification [A]. International Workshop on Recognition, Analysis and Tracking of Faces and Gestures in Real-Time Systems [C], 1999: 58-63.
[25] L.Jordao, M.Perrone, J.Costeira, J.Santos-Victor.Active Face and Feature Tracking [A]. Proceedings of the 10th International Conference on Image Analysis and Processing [C], 1999: 572-577.
[26] L.Sigal, S.Sclaroff,V Athitsos.Estimation and Prediction of Evolving Color Distributions for Skin Segmentation under Varying Illumination [A]. Proceeding of IEEE Conference on Computer Vision and Pattern Recognition [C], 2000, 2:152-159.
[27] S.Mckenna, S.Gong,Y Raja.Modeling Facial Colour and Identity with Gaussian Mixtures[J].Pattern Recognition,1999,31(12):1883-1892.
[28]Visual C++数字图像模式识别技术及工程实践[M].北京:人民邮电出版社,2003年2月版.
[29]J.Kovae. P. Peer, F. Solina. Human Skin Colour Clustering for Face Detection[A].E- UROCON 2003, Computer as a Tool[C],2003:144-148.
[30]L.Jordao, M.Perrone, J.Costeira, J.Santos-Victor. Active Face and Feature Tracking [A].Proceedings of the 10th International Conference on Image Analysis and Processing [C], 1999: 572-577.
[31]J.Ahlberg.A System for Face Localization and Facial Feature Extraction [R]Techni- cal Report, LiTH-ISY R-2172, Linkoping University, 1999.
[32]R-L.Hsu,A-M.Mohamed,A.K.,Jain.Face Deteetion in Color Images[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(5):696-706.
[33]J.Ahlberg.A System for Face Localization and Facial Feature Extraction [R].Technical Report,LiTH-ISY-R-2172,Linkoping Univerity,1999.
[34] M.Yang,N.Ahuja. Gaussian mixture model for human skin color and its application in image and video databases[A].Proceedings of the SPIE: Conference on Storage and Retrieval for Image and Video Databases[C],1999,3656:458-466.
[35]M.J.JONES , J.M.REHG. Statistical color models with application to skin detection[A]. Proeeedings of IEEE Conference on Computer Vision and Pattern Recognition[C],1999:274-280.
[36] M.Yang,N.Ahuja. Gaussian mixture model for human skin color and its application in image and video databases[A].Proceedings of the SPIE: Conference on Storage and Retrieval for Image and Video Databases[C],1999,3656:458-466.
[37]J-C.Terrillon,M.N.Shirazi,H.Fukamaehi,S.Akamatsu. Comparative Performance of Differeni Skin Chrominance Models and Chrominance Spaces for the AutomaticDetection of Human Faces in Color Images[A].Proceeding of the International Conference on Face and Gesture Recognition[C」,2000,54一61.
[38]N.Oliver,A.Pentland,F.Berard.Lafter:Lips and Face Real Time Tracker with Facial Expression Recognition[A].Proeeedings of Computer Vision and Pattern Recognition[c] 1997:123-129.
[39]田捷,杨鑫.生物特征识别技术理论与应用[M].2005,北京,电子工业出版社.
[40] Rafael C. Gonzalez Richad E. Woods.数字图像处理(第一版)[M].阮秋琦,阮宇智.北京.电于工业出版社,2003.
[41]Hsu Rdirr-Lien,Mohamed,Jain Anil K. Face detection in color images[J].IEEE Trans on Pattern Analysis and Machine Intelligence,2002,24(5):696-706.
[42]《A System for human face and facial feature location》.
[43]Visual C++数字图像模式识别技术及工程实践[M].北京:人民邮电出版社,2003年2月版.
[44]汪更生,林家骏,刘云翔,基于色度信息及投影法的彩色图像中人脸区域分割方法,长春工程学院学报(自然科学版),2004,5(3):51一54.
[45]E. Osuna. Support Vector Machines: Training and Applications.PhD thesis, MIT, EE/CS Dept., Cambridge, MA,1998.
[46]Visual C++数字图像模式识别技术及工程实践[M].北京:人民邮电出版社,2003年2月版.
[47]张志刚,周明全,耿国华.人脸关键特征点自动标定研究[J].计算机工程与应用.2007,43(21):197-217.
[48]田捷,杨鑫.生物特征识别技术理论与应用[M].2005,北京,电子工业出版社.
[49]朱夏君。人脸识别的研究及眼睛定位算法〔C],硕士学位论文,浙江工业大学,2004
[50]段立娟,崔国勤,高文等,多层次特定图像过滤方法[J].计算机辅助设计与图形学报,2002,14(5):404409.
[51]衣晓飞,陈福接。单人图像中人眼的快速定位[[J],计算机工程与科学,2001 23 (3)
[52]石东海。人脸检测和定位方法研究[C],硕士学位论文,国防科学技术大学,2002 1
[53]冈萨雷斯。数字图像处理,第二版。电子工业出版社
[54]邓加琼.复杂背景下的彩色人脸检测算法与实现「D]。成都:电子科技大学,2005. 19-21
[55]王良民,张建明,詹永照,宋顺林。人脸检测研究现状和发展。江苏大学学报(自然科学版)第24卷第3期2003年5月
[56]唐伟,陈兆乾,吴建鑫,周志华。静态图像中的人脸检测方法综述。计算机科学,2002年20期