Abstract
In this paper, a novel classifier named Kernel Nearest-Farthest Subspace (KNFS) classifier is proposed for face recognition. Inspired by the kernel-based classifier and the Nearest-Farthest Subspace (NFS) classifier, KNFS can make the sample points to be linear separable by utilizing the kernel function to map linear inseparable sample points in low-dimensional space to high-dimensional kernel space. And it can improve the recognition accuracy of crossed sample points between classes. The algorithm provides the highest reported recognition accuracy on AR and AT&T database. The results are comparable with many other state-of-art face recognition algorithms.
Original language | English |
---|---|
Pages (from-to) | 32007-32021 |
Number of pages | 15 |
Journal | Multimedia Tools and Applications |
Volume | 78 |
Issue number | 22 |
DOIs | |
State | Published - Nov 1 2019 |
Keywords
- Face recognition
- Kernel function
- Nearest-farthest subspace classifier