NMF Face Recognition Method Based on Alpha Divergence

Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 217)


This paper proposed NMF decomposition method based on Alpha divergence for face recognition, which uses Alpha divergence as a distance measure standard to obtain the corresponding NMF decomposition expression. Through the parameter values derived from the expression, a variety of decomposition iteration expression can be obtained. In each iteration process, the differences are calculated to determine the optimal parameters of the next step. Such decomposition can converge to the global optimum to improve the accuracy of face recognition.


NMF Alpha divergence Face recognition 


  1. 1.
    Moon H, Phillips PJ (2001) Computational and performance aspects of PCA-based face recognition algorithms. Perception 30:303–321CrossRefGoogle Scholar
  2. 2.
    Lu j, Plataniotis KN, Venetsanopoulos AN (2003) Face recognition using kernel direct discriminant analysis algorithms. IEEE Trans Neural Networks 14(1):117–126Google Scholar
  3. 3.
    Bartlett MS, Movellan JR, Sejnowski TJ (2002) Face recognition by independent component analysis. IEEE Trans Neural Networks 13(6):450–1464CrossRefGoogle Scholar
  4. 4.
    Dhillon S, Sra S (2005) Generalized nonnegative matrix approximations with Bregman divergences. Adv Neural Inf Process Syst 05:19–24Google Scholar
  5. 5.
    Lin C-J (2007) Projected gradient methods for nonnegative matrix factorization. Neural Comput 19:2756–2779MathSciNetMATHCrossRefGoogle Scholar
  6. 6.
    Amari S, Cichocki A (2010) Information geometry of divergence functions. Bull Pol Acad Sci 58:183–195Google Scholar
  7. 7.
    Cichocki A, Zdunek R, Phan AH, Amari S (2009) Nonnegative matrix and tensor factorizations, vol 47. Wiley, Chichester, pp 24–27Google Scholar

Copyright information

© Springer-Verlag London 2013

Authors and Affiliations

  1. 1.Chongqing City Management CollegeChongqingChina

Personalised recommendations