结合伽马变换和小波变换的PCA人脸识别算法
|
摘要
为了有效地提取人脸特征,提出了一种在传统PCA算法的基础上,结合伽马变换与小波变换的人脸识别算法。该方法对人脸图像进行伽马变换,消除光照等非线性因素的影响;对变换后的人脸图像进行小波分解,用得到的低频分量来替代原始人脸;对得到的人脸低频分量作PCA特征提取,得到最终的鉴别特征。在ORL人脸库上进行测试,该算法的识别率比传统的PCA算法提高了6.5%。
In order to extract the features of human faces effectively, a face recognition algorithm based on traditional PCA method is proposed in this article, which combined with gamma transform and wavelet transform. The face images are processed with gamma transform, which can eliminate the effects of light and other nonlinear factors. It decomposes the face images by wavelet transform. It uses the low frequency component to instead the original face. It extracts face feature from the low frequency component of face by PCA to get the final identification characteristics. When tested in ORL faces database, the recognition rate of this algorithm is 6.5% higher than traditional algorithm based on PCA.
In order to extract the features of human faces effectively, a face recognition algorithm based on traditional PCA method is proposed in this article, which combined with gamma transform and wavelet transform. The face images are processed with gamma transform, which can eliminate the effects of light and other nonlinear factors. It decomposes the face images by wavelet transform. It uses the low frequency component to instead the original face. It extracts face feature from the low frequency component of face by PCA to get the final identification characteristics. When tested in ORL faces database, the recognition rate of this algorithm is 6.5% higher than traditional algorithm based on PCA.
引文
[1]焦斌亮,陈爽.基于PCA算法的人脸识别[J].计算机工程与应用,2011,47(18):201-203.
[2]李盛文,鲍苏苏.基于PCA+Ada Boost算法的人脸识别技术[J].计算机工程与应用,2010,46(4):170-173.
[3]Turk M A,Pentland A P.Face recognition using eigenfaces[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition,1991:586-591.
[4]Yang J,Zhang D,Frangi A F,et al.Two-dimensional PCA:a new approach to appearance-based face representation and recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2004,26(1):131-137.
[5]Yang Jian,Yang Jingyu.From image vector to matrix:a straightforward image projection technique—IMPCA vs.PCA[J].Pattern Recognition,2002,35(9):1997-1999.
[6]Wang L,Wang X,Zhang X,et al.The equivalence of twodimensional PCA to line-based PCA[J].Pattern Recognition Letters,2005,26(1):57-60.
[7]Zhang D,Zhou Z H.(2D)2PCA:two-directional twodimensional PCA for efficient face representation and recognition[J].Neurocomputing,2005,69(1):224-231.
[8]陈粟,倪林.一种特征脸分析和小波变换相结合的人脸识别方法[J].计算机应用,2004,24(10):75-77.
[9]Cheng Y,Wang C,Zhao C.Illumination and posture invariant face recognition using Radon and nonsubsampled contourlet transform[C]//Chinese Conference on Pattern Recognition,2009:1-5.
[10]Chen W,Er M J,Wu S.Illumination compensation and normalization for robust face recognition using discrete cosine transform in logarithm domain[J].IEEE Transactions on Systems,Man,and Cybernetics,Part B:Cybernetics,2006,36(2):458-466.
[11]黄静.基于光照预处理的人脸检测研究与应用[D].黑龙江大庆:东北石油大学,2012.
[12]Gonzalez R C,Woods R E,Eddins S L.数字图像处理:MATLAB版[M].北京:电子工业出版社,2005.
[13]聂会星,梁坤,徐枞巍.基于小波变换和支持向量机的人脸识别研究[J].合肥工业大学学报:自然科学版,2011,34(2):208-211.
[14]王科俊,邹国锋.基于子模式的Gabor特征融合的单样本人脸识别[J].模式识别与人工智能,2013,26(1):50-56.
[15]周国民,陈勇,李国军.人脸识别中应用小波变换的两个关键问题[J].浙江大学学报:理学版,2005,32(1):34-38.
[2]李盛文,鲍苏苏.基于PCA+Ada Boost算法的人脸识别技术[J].计算机工程与应用,2010,46(4):170-173.
[3]Turk M A,Pentland A P.Face recognition using eigenfaces[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition,1991:586-591.
[4]Yang J,Zhang D,Frangi A F,et al.Two-dimensional PCA:a new approach to appearance-based face representation and recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2004,26(1):131-137.
[5]Yang Jian,Yang Jingyu.From image vector to matrix:a straightforward image projection technique—IMPCA vs.PCA[J].Pattern Recognition,2002,35(9):1997-1999.
[6]Wang L,Wang X,Zhang X,et al.The equivalence of twodimensional PCA to line-based PCA[J].Pattern Recognition Letters,2005,26(1):57-60.
[7]Zhang D,Zhou Z H.(2D)2PCA:two-directional twodimensional PCA for efficient face representation and recognition[J].Neurocomputing,2005,69(1):224-231.
[8]陈粟,倪林.一种特征脸分析和小波变换相结合的人脸识别方法[J].计算机应用,2004,24(10):75-77.
[9]Cheng Y,Wang C,Zhao C.Illumination and posture invariant face recognition using Radon and nonsubsampled contourlet transform[C]//Chinese Conference on Pattern Recognition,2009:1-5.
[10]Chen W,Er M J,Wu S.Illumination compensation and normalization for robust face recognition using discrete cosine transform in logarithm domain[J].IEEE Transactions on Systems,Man,and Cybernetics,Part B:Cybernetics,2006,36(2):458-466.
[11]黄静.基于光照预处理的人脸检测研究与应用[D].黑龙江大庆:东北石油大学,2012.
[12]Gonzalez R C,Woods R E,Eddins S L.数字图像处理:MATLAB版[M].北京:电子工业出版社,2005.
[13]聂会星,梁坤,徐枞巍.基于小波变换和支持向量机的人脸识别研究[J].合肥工业大学学报:自然科学版,2011,34(2):208-211.
[14]王科俊,邹国锋.基于子模式的Gabor特征融合的单样本人脸识别[J].模式识别与人工智能,2013,26(1):50-56.
[15]周国民,陈勇,李国军.人脸识别中应用小波变换的两个关键问题[J].浙江大学学报:理学版,2005,32(1):34-38.