Sparse Discriminant Preserving Projections for Face Recognition

Jianbo Zhang, Jinkuan Wang, Kun Zhang


Previous works have demonstrated that dimensionality reduction algorithms that combine sparse subspace learning (SSL) and discriminant information of sample data can improve the classification performance for some pattern recognition problems. However, most of these approaches introduce within-class sparse reconstruction matrix and within-class discriminative structure simultaneously, which may affect each other. To address this problem, in this paper, we propose a new dimensionality reduction algorithm called sparse discriminant preserving projections (SDPP).Different from the existing methods, SDPP uses between-class scatter as global discriminant information and integrates it with sparse reconstructive structure of samples in each class to establish the objective function. Since SDPP only takes into account the within-class sparsity reconstructive relationship and the between-class scatter information, it not only well preserves the sparse reconstruction relationship, but also improves classification performance. Experiments on face image databases demonstrate the superiority of the proposed algorithm.


Dimensionality Reduction, Sparse Representation, Sparse Subspace Learning, Face Recognition


Full Text:



  • There are currently no refbacks.