Research Article
An Efficient Feature Extraction Method, Global Between Maximum and Local Within Minimum, and Its Applications
Table 3
The best recognition accuracy (%) of different algorithms and the numbers in the bracket corresponding to standard deviation on the ORL database.
| Size | PCA | LDA | MMC | LPP | NPE | ANMM | GBMLWM |
|
3
| 77.43 (0.03) | 87.57 (0.03) | 84.61 (0.02) | 70.04 (0.04) | 80.86 (0.03) | 84.54 (0.03) | 90.07 (0.03) |
4
| 84.71 (0.02) | 91.08 (0.02) | 90.96 (0.02) | 74.33 (0.03) | 87.04 (0.02) | 91.25 (0.02) | 94.64 (0.02) |
5
| 88.20 (0.02) | 93.95 (0.02) | 93.95 (0.02) | 80.15 (0.03) | 90.45 (0.02) | 94.15 (0.02) | 97.20 (0.02) |
6
| 89.44 (0.03) | 94.37 (0.02) | 94.56 (0.02) | 81.69 (0.03) | 91.56 (0.03) | 95.12 (0.03) | 97.31 (0.02) |
|
|