基于核的正交局部保持投影的人脸识别

金一; 阮秋琦

doi:10.3724/SP.J.1146.2007.01450

基于核的正交局部保持投影的人脸识别

doi: 10.3724/SP.J.1146.2007.01450

金一,
阮秋琦

基金项目:

国家973计划项目(2004CB318005)和国家自然科学基金(60472033，60672062)资助课题

计量
- 文章访问数: 4079
- HTML全文浏览量: 111
- PDF下载量: 1400
- 被引次数: 0
出版历程
- 收稿日期: 2007-09-11
- 修回日期: 2007-12-24
- 刊出日期: 2009-02-19

Kernel Based Orthogonal Locality Preserving Projections for Face Recognition

Jin Yi,
Ruan Qiu-qi

摘要

摘要: 针对发掘人脸图像中的高维非线性结构，本文将加核及向量间相互正交两种思想同时引入局部保留投影算法中，提出了一种新的基于核的正交局部保持投影(Kernel based Orthogonal Locality Preserving Projections, KOLPP)的非线性子空间人脸识别算法并给出了其推导过程。该算法首先利用核的方法提取人脸图像中的非线性信息，并将其投影在一个高维非线性空间，在保证各向量正交的同时，通过局部保持投影算法做一线性映射，从而更好地提取人脸非线性局部邻域结构特征。在ORL和Yale人脸库上的试验证明了该文所提算法的有效性。
- 人脸识别;特征提取;核方法;正交局部保持投影
Abstract: In this paper, considering kernel and orthogonal basis functions, a new method named kernel based orthogonal locality preserving projections algorithm, which aims at discovering an embedding that preserves nonlinear information is proposed for face representation and recognition. In this algorithm, first, the nonlinear kernel mapping is used to map the face data into an implicit feature space, and then a linear transformation which produces orthogonal basis functions is performed to preserve locality geometric structures of the face image. Experiments based on both ORL and Yale face database demonstrate the effectiveness of the new algorithm.

HTML全文

参考文献(1)

Turk M and Pentland A. Eigenfaces for recognition. Journalof Cognitive Neuroscience, 1991, 3(1): 72-86.[2]Belhumeur P N, Hespanha J P, and Kriegman D J.Eigenfaces vs. fisherfaces: Recognition using class specificlinear projection. IEEE Trans. on Pattern Analysis andMachine Intelligence, 1997, 19(7): 711-720.[3]Bengio Y, Palement J, and Vincent P, et al.. Out-of-sampleextensions for LLE, Isomap, MOS, eigenmaps, and spectralclustering. Advances in Neural Information ProcessingSystems 6. Cambridge, MA: MIT Press, 2003: 177.[4]Roweis S T and Saul L K. Nonlinear dimensionality reductionby locally linear embedding[J].Science.2000, 290(5500):2323-2326[5]Belkin M and Niyogi P. Laplacian eigenmaps and spectraltechniques for embedding and clustering. Advances in NeuralInformation Processing Systems 14, Vancouver: MIT Press,2001: 585-591.[6]Yang M H, Ahuja N, and Kriegman D. Face recognition usingkernel Eigenfaces. Proceedings of International Conference onImage Processing, Vancouver, Canada, September 2000:37-40.[7]Scholkopf B, Smola A, and Muller K R. Nonlinear componentanalysis as a kernel eigenvalues problem[J].Neural Computation.1998, 10(5):1299-1319[8]He Xiaofei and Yan Shuicheng, et al.. Face recognition usinglaplacianfaces[J].IEEE Trans. on Pattern Analysis and MachineIntelligence.2005, 27(3):328-340[9]Deng Cai, He Xiaofei, and Han Lianwei, et al.. OrthogonalLaplacianfaces for face recognition. IEEE Trans. on ImageProcessing, 2006, 15(11): 3608-3614.[10]Cheng J, Liu Q S, and Lu H Q, et al.. Supervised kernellocality preserving projections for face recognition[J].Neurocomputing Letter.2005, 67:443-449[11]祝磊, 朱善安. KSLPP：新的人脸识别算法. 浙江大学学报(工学版), 2007, 41(41): 1066-1069.[12]万海平. 模式识别中核方法若干问题研究. [博士论文], 北京邮电大学, 2005.

施引文献

资源附件(0)

访问统计