• Face Recognition Based on Sparsity Preserving Pro- jections and Gabor Transform   [MASS 2012]
  • Author(s)
  • Xinwu Yang, Xiaojun Xu, Chunnian Liu
  • Sparsity preserving projections (SPP) rooted in com- pressive sensing has already been successfully applied in face recognition. But we can only get good recognition results on a large dataset by SPP. It is difficult to get similar high recognition rate on a small dataset. However, in many practical applications, we have to use SPP on small-scale datasets. In order to obtain a better recognition rate on a small number of samples, we pro- posed a new face recognition method integrating Gabor trans- form and SPP (GTSPP). In the new method, we firstly carry out Gabor transform on every face sample at a combination of dif- ferent frequency and direction. Then we constitute a new dataset that consists directly of the coefficients of Gabor transform at each combination of frequency and direction. The new dataset is larger than the original dataset. Finally, SPP is used to recognize human identity on the new larger dataset. We conduct experi- ments on publicly available dataset (Yale and AR) to verify the feasibility and effectiveness of the proposed method and explain why it has a higher recognition rate than SPP.
  • Gabor transform; sparsity preserving projections; compressive sensing; face recognition
  • References
  • [1]
    Sun Dongmei and Qiu Zhengding, "A survey of the emerging biometric technology," ACTA ELECTRONICE SINICA, vol.29, pp.1744-1748, 2001.
    L. S. Qiao, S. C. Chen and X. Y. Tan, "Sparsity preserving projections with applications to face recognition," Pattern Recognition, vol.43, pp. 331-341, 2010.
    R. G. Baraniuk, "A lecture of Compressive sensing," IEEE Signal Processing Magazine, vol.24, pp. 118-120, 2007.
    Alan C.Bovic A C, Clark M, Geisler W S, "Multichannel texture analy- sis using localized spatial filters," IEEE Transactions on Pattern Analy- sis and Machine Intelligence, vol.12, pp.55-73, 1990.
    Zeng Shuyan,Zhang Guangjun and Li Xiuzhi, "Image target distinguish based on Gabor filters," Chinese journal of Beijing University of Aero- nautics and Astronautics, vol.32, 2006.
    Wang Huiyuan, "Face recognition approaches based on linear subspace and circularly symmetrical Gabor transform," Chinese journal of Shandong University Doctoral Dissertation, pp:53-57, 2007.
    D.Donoho, "Compressed sensing, " IEEE Transaction on Information Theory,vol. 52, pp.1289-1306, 2006.
    E. Candes and J. Romberg, "Sparsity and incoherence in compressive sampling," Inverse Problems, vol.23, pp. 969-985, 2007.
    Stephen Boyed and Lieven Vandenberghe, "Convex Optimization," Cambridge University Press,2004.
    Emmanuel Candes and Justin Romberg,"l 1 -MAGIC: Recovery of sparse signals via convex programming," 2005,unpublished.
    J. Wright, A. Y. Yang, A. Ganesh, S. S. Sastry and Y. Ma, "Robust face recognition via sparse representation," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol.31, pp. 210-227, 2009.
    Wei Na, Geng Guohua and Zhou Mingquan, "Content-based image retrieval using Gabor filters," Chinese journal of Computer Engineering, vol.31, 2005.

Engineering Information Institute is the member of/source content provider to