Conference Paper

Face Recognition via Two Dimensional Locality Preserving Projection in Frequency Domain.

DOI: 10.1016/j.neucom.2011.08.045 Conference: Life System Modeling and Intelligent Computing - International Conference on Life System Modeling and Simulation, LSMS 2010, and International Conference on Intelligent Computing for Sustainable Energy and Environment, ICSEE 2010, Wuxi, China, September 17-20, 2010. Proceedings, Part III
Source: DBLP

ABSTRACT In this paper we propose a new face recognition method based on two-dimensional locality preserving projections (2DLPP) in frequency domain. For this purpose, we first introduce the two-dimensional locality preserving projections. Then the 2DLPP in frequency domain is proposed for face recognition. In fact, two dimensional discrete cosine transform (2DDCT) is used as a pre-processing step and it transforms the face image signals from spatial domain into frequency domain aiming to reduce the effects of illumination and pose changes in face recognition. Then 2DLPP is applied on the upper left corner blocks of the 2DDCT transformed matrices, which represent main energy of each original image. For demonstration, the Olivetti Research Laboratory (ORL), YALE, FERET and YALE-B face datasets are used to compare the proposed approach with the conventional 2DLPP and 2DDCT approaches with the nearest neighborhood (NN) classifier. The experimental results show that the proposed 2DLPP in frequency domain is superior over the 2DLPP in spatial domain and 2DDCT itself in frequency domain.

  • [Show abstract] [Hide abstract]
    ABSTRACT: We consider the problem of locality preserving projections (LPP) in two-dimensional sense. Recently, LPP was proposed for dimensionality reduction, which can detect the intrinsic manifold structure of data and preserve the local information. As far as matrix data, such as images, are concerned, they are often vectorized for LPP algorithm to find the intrinsic manifold structure. While the dimension of matrix data is usually very high, LPP cannot be implemented because of the singularity of matrix. In this paper, we propose a method called two-dimensional locality preserving projections (2D-LPP) for image recognition, which is based directly on 2D image matrices rather than 1D vectors as conventional LPP does. From an algebraic procedure, we induce that 2D-LPP is related to two other linear projection methods, which are based directly on image matrix: 2D-PCA and 2D-LDA. 2D-PCA and 2D-LDA preserve the Euclidean structure of image space, while 2D-LPP finds an embedding that preserves local information and detects the intrinsic image manifold structure. To evaluate the performance of 2D-LPP, several experiments are conducted on the ORL face database, the Yale face database and a digit dataset. The high recognition rates and speed show that 2D-LPP achieves better performance than 2D-PCA and 2D-LDA. Experiments even show that conducting PCA after 2D-LPP achieves higher recognition than LPP at the same dimension of feature spaces.
    Neurocomputing 01/2007; 70:912-921. · 2.01 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, a new technique coined two-dimensional principal component analysis (2DPCA) is developed for image representation. As opposed to PCA, 2DPCA is based on 2D image matrices rather than 1D vectors so the image matrix does not need to be transformed into a vector prior to feature extraction. Instead, an image covariance matrix is constructed directly using the original image matrices, and its eigenvectors are derived for image feature extraction. To test 2DPCA and evaluate its performance, a series of experiments were performed on three face image databases: ORL, AR, and Yale face databases. The recognition rate across all trials was higher using 2DPCA than PCA. The experimental results also indicated that the extraction of image features is computationally more efficient using 2DPCA than PCA.
    IEEE Transactions on Pattern Analysis and Machine Intelligence 02/2004; · 5.69 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper we propose the discrete cosine transform (DCT) feature set, which utilizes polynomial coefficients derived from 2D DCT coefficients obtained from spatially neighboring blocks. Face verification results on the multi-session VidTIMIT database suggest that the DCT- feature set is superior (in terms of robustness to illumination direction changes and discrimination ability) to features extracted using three popular methods: eigenfaces principal component analysis, 2D DCT and 2D Gabor wavelets. Moreover, compared to Gabor wavelets, the DCT- feature set is over 80 times faster to compute. Additional experiments on the Weizmann database also show that the DCT- approach is more robust than 2D Gabor wavelets and 2D DCT coefficients.
    Signal Processing 05/2003; · 2.24 Impact Factor