Research Article
An Efficient Feature Extraction Method, Global Between Maximum and Local Within Minimum, and Its Applications
Table 2
Recognition accuracy (%) of different algorithms and the numbers in the bracket corresponding to dimensions on the UMIST database.
| Size | PCA | LDA | MMC | LPP | NPE | ANMM | GBMLWM |
|
3
| 77.26 (50) | 77.55 (17) | 80.06 (44) | 71.61 (32) | 70.60 (38) | 77.90 (17) | 79.88 (15) |
4
| 76.08 (67) | 82.73 (19) | 83.94 (67) | 75.43 (50) | 75.74 (60) | 81.60 (22) | 86.10 (13) |
5
| 82.69 (87) | 87.73 (19) | 90.59 (77) | 81.28 (77) | 81.43 (80) | 88.48 (20) | 91.85 (11) |
6
| 85.72 (68) | 89.52 (13) | 92.81 (96) | 85.74 (77) | 85.82 (100) | 91.91 (13) | 93.80 (18) |
|
|