Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
531511 | Pattern Recognition | 2008 | 10 Pages |
Abstract
In this paper, we propose a new linear subspace analysis algorithm, called orthogonal neighborhood preserving discriminant analysis (ONPDA). Given a set of data points in the ambient space, a weight matrix is firstly built which describes the relationship between the data points. Then optimal between-class scatter matrix and within-class scatter matrix are defined such that the neighborhood structure can be preserved. In order to improve the discriminating power, a new method is presented for orthogonalizing the basis eigenvectors. We evaluate the performance of the proposed algorithm for face recognition with the use of different databases. Consistent and promising results demonstrate the effectiveness of our algorithm.
Keywords
Related Topics
Physical Sciences and Engineering
Computer Science
Computer Vision and Pattern Recognition
Authors
Haifeng Hu,