人脸图像光照预处理算法研究
|
摘要
人脸识别作为一种主要的身份识别手段,在公共安全、视觉监控、数字身份认证、电子商务、多媒体和数字娱乐等领域有着广泛的应用前景。经过近四十年的发展,人脸识别技术已经取得了长足的进步,基本实现特定环境下的准确识别,并陆续出现了一些优秀的商用识别软件。但是在环境不可控的情况下,由于光照、姿态、表情、遮挡等变化的影响,已有的人脸识别算法性能大大下降,其应用范围受到较大的限制。其中,以光照问题给人脸识别带来的影响尤为显著。相关研究表明,同一个人的人脸图像在光照条件不同和经过各种主流的人脸识别方法特征提取后引起的差异,往往要大于不同的人在相同光照条件下的人脸图像的差异。因此,如何有效地对人脸图像进行光照纠正,以达到光照无关的预处理效果,是人脸识别研究的一个重要课题。为此,本文做了以下工作：
(1)分析基于Retinex理论的光照预处理算法,并对算法进行改进：要消除光照对人脸图像的影响,在对数域执行高通滤波之前,先利用小波变换对图像进行高频强化。
(2)为了提高人脸预处理识别率,将基于Retinex理论的W-G算法、直方图均衡化、gamma灰度校正三种预处理算法结合,寻找一种最优的组合方式。
(3)分别使用PCA、LDA识别算法对上述光照补偿算法的预处理结果进行识别,结合各种光照补偿算法的特点对识别的结果进行分析,人脸数据库使用Yale B人脸数据库。比较分析两种识别方法的识别时间、识别率。
实验结果发现,文中设计的五种光照预处理算法,在光照不理想的情况下均能取得较好的识别率；在正常光照情况下,不会引起图像质量的恶化。其中均衡化与W-G算法的分步处理得到的预处理效果最佳。直方图均衡化对灰度的均衡作用,削弱滤波中引起的光晕效果,同时高斯滤波滤去图像的低频分量,即与光照有关的入射分量,消除光照变化引起的图像的差异。
As one of the most successful branches of personal identification, face recognition has great potential applications in public safely, visual surveillance, digital identity certification, e-commerce, multimedia and digital entertainment and other fields. The face recognition technology has developed rapidly over the past 40 years, basically accurate identification of a specific environment, and has been found in some excellent commercial recognition software. While the face recognition algorithm performance degrades under uncontrolled environment such as variant illumination, head poses, facial expressions, occlusion on face and so on. Research indicates that, in face recognition, variations caused by illumination are more significant than the inherent differences between individuals. Therefore, how to effectively correct the illumination face images to achieve the pre-independent effect of light is an important topic in Face Recognition.
In our study paper, the main work can be summarized as follows:
(1) Studies the illumination preprocessing algorithm based on Retinex theory, improve the algorithm as:to remove the effect of illumination on human face images, enhancing the high-frequency of the image using wavelet transform before the implementation of the log-domain high-pass filter with the image.
(2) In order to improve recognition rate of face image preprocess, combining three preprocessing algorithms with:improved light compensation algorithm based on Retinex theory, histogram equalization and gray-scale transformation of gamma, find an optimal combination.
(3) Identify the pre-processing results of illumination compensation algorithm with PCA, LDA recognition algorithm, using the Yale B face database. Combine the characteristics of illumination compensation algorithm to analyze the results of the identification, Comparative of two identification methods with identification time, recognition rate.
The results showed that the five light preprocessing algorithm design in the test, achieve better recognition rate when light environment is not ideal; in normal lighting circumstances, the preprocessing algorithm does not cause the deterioration of image quality. Which equalization and improved illumination compensation algorithm sub-step processing is best handled by the pretreatment. Histogram equalization, balanced on the role of gray scale, weakening the halo effect caused by homomorphic filtering, and homomorphic filtering filters the low frequency components of image, eliminate the image differences caused by illumination change.
(1)分析基于Retinex理论的光照预处理算法,并对算法进行改进：要消除光照对人脸图像的影响,在对数域执行高通滤波之前,先利用小波变换对图像进行高频强化。
(2)为了提高人脸预处理识别率,将基于Retinex理论的W-G算法、直方图均衡化、gamma灰度校正三种预处理算法结合,寻找一种最优的组合方式。
(3)分别使用PCA、LDA识别算法对上述光照补偿算法的预处理结果进行识别,结合各种光照补偿算法的特点对识别的结果进行分析,人脸数据库使用Yale B人脸数据库。比较分析两种识别方法的识别时间、识别率。
实验结果发现,文中设计的五种光照预处理算法,在光照不理想的情况下均能取得较好的识别率；在正常光照情况下,不会引起图像质量的恶化。其中均衡化与W-G算法的分步处理得到的预处理效果最佳。直方图均衡化对灰度的均衡作用,削弱滤波中引起的光晕效果,同时高斯滤波滤去图像的低频分量,即与光照有关的入射分量,消除光照变化引起的图像的差异。
As one of the most successful branches of personal identification, face recognition has great potential applications in public safely, visual surveillance, digital identity certification, e-commerce, multimedia and digital entertainment and other fields. The face recognition technology has developed rapidly over the past 40 years, basically accurate identification of a specific environment, and has been found in some excellent commercial recognition software. While the face recognition algorithm performance degrades under uncontrolled environment such as variant illumination, head poses, facial expressions, occlusion on face and so on. Research indicates that, in face recognition, variations caused by illumination are more significant than the inherent differences between individuals. Therefore, how to effectively correct the illumination face images to achieve the pre-independent effect of light is an important topic in Face Recognition.
In our study paper, the main work can be summarized as follows:
(1) Studies the illumination preprocessing algorithm based on Retinex theory, improve the algorithm as:to remove the effect of illumination on human face images, enhancing the high-frequency of the image using wavelet transform before the implementation of the log-domain high-pass filter with the image.
(2) In order to improve recognition rate of face image preprocess, combining three preprocessing algorithms with:improved light compensation algorithm based on Retinex theory, histogram equalization and gray-scale transformation of gamma, find an optimal combination.
(3) Identify the pre-processing results of illumination compensation algorithm with PCA, LDA recognition algorithm, using the Yale B face database. Combine the characteristics of illumination compensation algorithm to analyze the results of the identification, Comparative of two identification methods with identification time, recognition rate.
The results showed that the five light preprocessing algorithm design in the test, achieve better recognition rate when light environment is not ideal; in normal lighting circumstances, the preprocessing algorithm does not cause the deterioration of image quality. Which equalization and improved illumination compensation algorithm sub-step processing is best handled by the pretreatment. Histogram equalization, balanced on the role of gray scale, weakening the halo effect caused by homomorphic filtering, and homomorphic filtering filters the low frequency components of image, eliminate the image differences caused by illumination change.
引文
[1]Alex Pentl, Baback Moghaddam, Thad starner. View-Based and Modular Eigen-spaces for Face Recognition. MIT Media Lab perceptual Computing Section Tech. IEEE Conference on Computer Vision&Pattern Recognition 1994
[2]A.Shashua and T.Riklin-Raviv. The Quotient Image:Class-Based Re-Rendering and Recognition With Varying Illuminations. IEEE Trans. Pattern Analysis and Machine Intelligence, vol.23, no.2, pp129-139,2001
[3]A.S.Georghiades, P.N.Belhumeur, D.J.Kriegman. From Few to Many:Illumination Cone Models for Face Recognition Under Variable Lighting and Pose. IEEE Trans. Pattern Analysis and Machine Intelligence, vol.23, no.6, pp643-660,2001
[4]A.S.Georghiades, D.J.Kriegman and P.N.Belhumeur, Illumination Cones for Recognition under Variable Lighting:Faces. Proceedings of International Conference on Computer Vision and Pattern Recognition, pp52-58,1998
[5]杜波.人脸识别中光照预处理研究.中国科学院研究生院硕士研究生论文.2005
[6]龚卫国,杨利平,辜小花,李伟红.基于多级小波分解的人脸图像光照补偿方法.光学精密工程.Vol.16,No.8,pp1459-1464.Aug 2008
[7]郭宇聪,张星明,詹皇源,张咏梅.基于多方法融合的人脸图像光照纠正算法.计算机工程与设计.Vol.27,No.9,ppl547-1559
[8]胡元奎,汪增福.可变光照条件下的人脸图像识别.中国图象图形学报.Vol.10,No.7,pp.844-849,July 2005
[9]H.Chan and W.W.Bledsoe, A machine facial recognition system:some preliminary results. Technical report, Panoramic Research Inc, Cal,1965
[10]琚生根,周激流,何坤,夏欣,王刚.频域光照归一化的人脸识别.电子科技大学学报.Vol.38,No.6,pp1021-1025,Nov.2009
[11]李粉兰,段海峰,郝建国,唐文彦.人脸识别中光照补偿问题的实验研究.工程图学学报.No.3,2009
[12]卢晓菁,陈锻生.一种人脸图像光照补偿的新方法.小型微型计算机系统.Vol.29,No.10,pp.1864-1867,2008
[13]李粉兰,唐文彦,段海峰.分数次幂多项式核函数在核直接判别式分析中的应用. 光学精密工程.pp1410-1414.Aug 2007
[14]罗振波,丁晓青,方驰.一种对光照鲁棒的人脸图像预处理算法.电视技术.pp78-80.Sep 2006
[15]卢晓菁.可变光照条件下人脸识别技术研究.华侨大学硕士学位论文.2007
[16]L.Qing,S.Shan, X.Chen. Face Relighting for Face Recognition under Generic Illumination. IEEE International Conference on Acoustics, Speech, and Signal Processing, pp733-736, May,2004
[17]L.Zhang, D.Samaras. Face Recognition under Variable Lighting using Harmonic Image Exemplars. Proceedings of International Conference on Computer Vision and Pattern Recognition, vol.Ⅰ, ppl9-25,2003
[18]M.Turk and A.Pentland, Eigenfaces for Recognition, Journal of cognitive neuroscience, vol.3, no. 1,pp71-86,1991
[19]聂祥飞,谭泽富,郭军.应用小波变换的人脸光照补偿.光学精密工程.Vol.16,No.1,pp.150-155，Jan.2008
[20]P.J.Philips, P.J.Grother, R.J.Micheals, D.M.Blackburn, E.Tabassi, and J.M.Bone, Face Recognition Vendor Test 2002:Evaluation report[R],NISTIR 6965, National Institute of Standards and Technology,2003, (http://www.frvt.org)
[21]P.J.Phillips, P.Grother, R.J.Micheals, D.M.Blackburn, E.Tabassi, and J.M.Bone “FRVT 2002:Evaluation Report”, http://www.frvt.org/DLs/FRVT_2002_Evaluation_Report.pdf, Mar. 2003
[22]P.J.Phillips, H.Moon,et al. The FERET Evaluation Methodology for Face-Recognition algorithms. IEEE Trans. Pattern Analysis and Machine Intelligence, vol.22, pp1019-1104, 2000
[23]P Belhumeur, J Hespanha, D Kriegman, Eigenfaces vs. Fisherfaces:Recognition using class specific linear projection. in Proceedings of Fourth European Conference on Computer Vision, ECCV'96, pp45-56,1996
[24]P N Bellhumer, J Hespanha, D Kriegman. Eigenfaces vs. fisherfaces:Recognition using class specific linear projection. IEEE Transactions on Pattern Analysis and Machine Intelligence, Special Issue on Face Recognition,17(7):711-720,1997
[25]P.Belhumeur and D.Kriegman, What Is the Set of Images of an Object under All Possible Illumination Conditions, International Journal of Computer Vision, vol.28, no.3, pp245-260, July 1998
[26]R.Basi and D.Jacobs. Lambertian Reflectance and Linear Subspaces. International Conference on Computer Vision, pp.383-390,2001
[27]Rafael C. Gonzalez, Richard E. Woods.数字图像处理(第二版)[M].北京电子工业出版社.Mar,2003
[28]R.Ramamoorthi and P.Hanrahan. On the Relationship between Radiance and Irradiance: Determining the Illumination from Images of a Convex Lambertian Object. Journal of the Optical Society of America A, vol.18, no.10, pp2448-2459,2001
[29]苏光大.计算机人像组合及人像自动查询系统走向应用,人民公安,no.1,pp.36-36,1999
[30]孙雪梅,苏菲,蔡安妮.人脸识别中光照补偿参数及必要性的判定方法.计算机应用研究.Vol.24,No.7,pp306-311.July,2007
[31]王蔚,王万良,宋加涛.基于小波的光照补偿及其在人脸识别中的应用.宁波大学学报(理工版).Vol.20,No.4,pp434-437.Dec,2007
[32]王彦臣,李树杰,黄廉卿.基于多尺度Retinex的数字X光图像增强方法研究.光学精密工程.pp70-76.Aug,2006
[33]温福喜,人脸识别技术研究[D].西安电子科技大学硕士学位论文,2007
[34]W.Zhao, R.Chellappa. SFS Based View Synthesis for Robust Face Recognition. Proceedings of the 4th International Conference on Face and Gesture Recognition, pp285-292, 2000
[35]W.Zhao and R.Chellappa, Illumination-insensitive Face Recognition using Symmetric Shape-from-Shading. Proceedings of International Conference on Computer Vision and Pattern Recognition, pp286-293,2000
[36]Xudong Xie, Kin-Man Lam. Face recognition under varying illumination based on a 2D face shape model. Pattern Recognition. Volume 38, Issue 2, February 2005, Pages 221-230
[37]Xudong Xie, Kin-Man Lam. An efficient illumination normalization method for face recognition. Pattern Recognition Letters. Volume 27,Issue 6,April 2006,Pages 609-617
[38]许高凤,丁士圻,郭丽华,王大成.基于小波的人脸去光照识别算法研究.系统仿真学报.Vol.21,No.14,pp.4362-4371.July,2009
[39]Young Kyung Park, Seok Lai Park, Joong Kyu Kim. Retinex method based on adaptive smoothing for illumination invariant face recognition. Signal Processing. Volume 88,Issue 8,August 2008,Pages 1929-1945
[40]Y.Adini, Y.Moses, S.Ullman. Face Recognition:the Problem of Compensating for Changes in Illumination Direction. IEEE Trans. Pattern Analysis and Machine Intelligence, vol.19, no.7,pp721-732,1997
[41]余磊.Gabor小波变换在人脸识别中的应用研究.重庆大学博士学位论文.2009.
[42]赵松,张志坚,张培仁.增强的典型相关分析及其在人脸识别特征融合中的应用.计算机辅助设计与图形学学报.Vol.21,No.3,pp.394-399.Mar,2009
[43]赵英男,孟宪权,徐勇,候春明.光照变化条件下的人脸识别预处理.吉首大学学报.Vol.28,No.2,pp.63-66,May.2007
[44]周激流,张晔.人脸识别理论研究进展.中国计算机辅助设计与图形学学报.Vol.12,No.2.Mar,1999
[45]International Biometric Group.Biometrics market and industry report 2009-2014[EB/OL]. 2009,http://www.biometricgroup.com.
[2]A.Shashua and T.Riklin-Raviv. The Quotient Image:Class-Based Re-Rendering and Recognition With Varying Illuminations. IEEE Trans. Pattern Analysis and Machine Intelligence, vol.23, no.2, pp129-139,2001
[3]A.S.Georghiades, P.N.Belhumeur, D.J.Kriegman. From Few to Many:Illumination Cone Models for Face Recognition Under Variable Lighting and Pose. IEEE Trans. Pattern Analysis and Machine Intelligence, vol.23, no.6, pp643-660,2001
[4]A.S.Georghiades, D.J.Kriegman and P.N.Belhumeur, Illumination Cones for Recognition under Variable Lighting:Faces. Proceedings of International Conference on Computer Vision and Pattern Recognition, pp52-58,1998
[5]杜波.人脸识别中光照预处理研究.中国科学院研究生院硕士研究生论文.2005
[6]龚卫国,杨利平,辜小花,李伟红.基于多级小波分解的人脸图像光照补偿方法.光学精密工程.Vol.16,No.8,pp1459-1464.Aug 2008
[7]郭宇聪,张星明,詹皇源,张咏梅.基于多方法融合的人脸图像光照纠正算法.计算机工程与设计.Vol.27,No.9,ppl547-1559
[8]胡元奎,汪增福.可变光照条件下的人脸图像识别.中国图象图形学报.Vol.10,No.7,pp.844-849,July 2005
[9]H.Chan and W.W.Bledsoe, A machine facial recognition system:some preliminary results. Technical report, Panoramic Research Inc, Cal,1965
[10]琚生根,周激流,何坤,夏欣,王刚.频域光照归一化的人脸识别.电子科技大学学报.Vol.38,No.6,pp1021-1025,Nov.2009
[11]李粉兰,段海峰,郝建国,唐文彦.人脸识别中光照补偿问题的实验研究.工程图学学报.No.3,2009
[12]卢晓菁,陈锻生.一种人脸图像光照补偿的新方法.小型微型计算机系统.Vol.29,No.10,pp.1864-1867,2008
[13]李粉兰,唐文彦,段海峰.分数次幂多项式核函数在核直接判别式分析中的应用. 光学精密工程.pp1410-1414.Aug 2007
[14]罗振波,丁晓青,方驰.一种对光照鲁棒的人脸图像预处理算法.电视技术.pp78-80.Sep 2006
[15]卢晓菁.可变光照条件下人脸识别技术研究.华侨大学硕士学位论文.2007
[16]L.Qing,S.Shan, X.Chen. Face Relighting for Face Recognition under Generic Illumination. IEEE International Conference on Acoustics, Speech, and Signal Processing, pp733-736, May,2004
[17]L.Zhang, D.Samaras. Face Recognition under Variable Lighting using Harmonic Image Exemplars. Proceedings of International Conference on Computer Vision and Pattern Recognition, vol.Ⅰ, ppl9-25,2003
[18]M.Turk and A.Pentland, Eigenfaces for Recognition, Journal of cognitive neuroscience, vol.3, no. 1,pp71-86,1991
[19]聂祥飞,谭泽富,郭军.应用小波变换的人脸光照补偿.光学精密工程.Vol.16,No.1,pp.150-155，Jan.2008
[20]P.J.Philips, P.J.Grother, R.J.Micheals, D.M.Blackburn, E.Tabassi, and J.M.Bone, Face Recognition Vendor Test 2002:Evaluation report[R],NISTIR 6965, National Institute of Standards and Technology,2003, (http://www.frvt.org)
[21]P.J.Phillips, P.Grother, R.J.Micheals, D.M.Blackburn, E.Tabassi, and J.M.Bone “FRVT 2002:Evaluation Report”, http://www.frvt.org/DLs/FRVT_2002_Evaluation_Report.pdf, Mar. 2003
[22]P.J.Phillips, H.Moon,et al. The FERET Evaluation Methodology for Face-Recognition algorithms. IEEE Trans. Pattern Analysis and Machine Intelligence, vol.22, pp1019-1104, 2000
[23]P Belhumeur, J Hespanha, D Kriegman, Eigenfaces vs. Fisherfaces:Recognition using class specific linear projection. in Proceedings of Fourth European Conference on Computer Vision, ECCV'96, pp45-56,1996
[24]P N Bellhumer, J Hespanha, D Kriegman. Eigenfaces vs. fisherfaces:Recognition using class specific linear projection. IEEE Transactions on Pattern Analysis and Machine Intelligence, Special Issue on Face Recognition,17(7):711-720,1997
[25]P.Belhumeur and D.Kriegman, What Is the Set of Images of an Object under All Possible Illumination Conditions, International Journal of Computer Vision, vol.28, no.3, pp245-260, July 1998
[26]R.Basi and D.Jacobs. Lambertian Reflectance and Linear Subspaces. International Conference on Computer Vision, pp.383-390,2001
[27]Rafael C. Gonzalez, Richard E. Woods.数字图像处理(第二版)[M].北京电子工业出版社.Mar,2003
[28]R.Ramamoorthi and P.Hanrahan. On the Relationship between Radiance and Irradiance: Determining the Illumination from Images of a Convex Lambertian Object. Journal of the Optical Society of America A, vol.18, no.10, pp2448-2459,2001
[29]苏光大.计算机人像组合及人像自动查询系统走向应用,人民公安,no.1,pp.36-36,1999
[30]孙雪梅,苏菲,蔡安妮.人脸识别中光照补偿参数及必要性的判定方法.计算机应用研究.Vol.24,No.7,pp306-311.July,2007
[31]王蔚,王万良,宋加涛.基于小波的光照补偿及其在人脸识别中的应用.宁波大学学报(理工版).Vol.20,No.4,pp434-437.Dec,2007
[32]王彦臣,李树杰,黄廉卿.基于多尺度Retinex的数字X光图像增强方法研究.光学精密工程.pp70-76.Aug,2006
[33]温福喜,人脸识别技术研究[D].西安电子科技大学硕士学位论文,2007
[34]W.Zhao, R.Chellappa. SFS Based View Synthesis for Robust Face Recognition. Proceedings of the 4th International Conference on Face and Gesture Recognition, pp285-292, 2000
[35]W.Zhao and R.Chellappa, Illumination-insensitive Face Recognition using Symmetric Shape-from-Shading. Proceedings of International Conference on Computer Vision and Pattern Recognition, pp286-293,2000
[36]Xudong Xie, Kin-Man Lam. Face recognition under varying illumination based on a 2D face shape model. Pattern Recognition. Volume 38, Issue 2, February 2005, Pages 221-230
[37]Xudong Xie, Kin-Man Lam. An efficient illumination normalization method for face recognition. Pattern Recognition Letters. Volume 27,Issue 6,April 2006,Pages 609-617
[38]许高凤,丁士圻,郭丽华,王大成.基于小波的人脸去光照识别算法研究.系统仿真学报.Vol.21,No.14,pp.4362-4371.July,2009
[39]Young Kyung Park, Seok Lai Park, Joong Kyu Kim. Retinex method based on adaptive smoothing for illumination invariant face recognition. Signal Processing. Volume 88,Issue 8,August 2008,Pages 1929-1945
[40]Y.Adini, Y.Moses, S.Ullman. Face Recognition:the Problem of Compensating for Changes in Illumination Direction. IEEE Trans. Pattern Analysis and Machine Intelligence, vol.19, no.7,pp721-732,1997
[41]余磊.Gabor小波变换在人脸识别中的应用研究.重庆大学博士学位论文.2009.
[42]赵松,张志坚,张培仁.增强的典型相关分析及其在人脸识别特征融合中的应用.计算机辅助设计与图形学学报.Vol.21,No.3,pp.394-399.Mar,2009
[43]赵英男,孟宪权,徐勇,候春明.光照变化条件下的人脸识别预处理.吉首大学学报.Vol.28,No.2,pp.63-66,May.2007
[44]周激流,张晔.人脸识别理论研究进展.中国计算机辅助设计与图形学学报.Vol.12,No.2.Mar,1999
[45]International Biometric Group.Biometrics market and industry report 2009-2014[EB/OL]. 2009,http://www.biometricgroup.com.