Research Article
A Robust and Fast Computation Touchless Palm Print Recognition System Using LHEAT and the IFkNCN Classifier
Table 4
Comparison of IFkNCN of different feature dimension values.
| Dim. | Clean | Salt and pepper | Motion blur | Time (s) | CA (%) | Time (s) | CA (%) | Time (s) | CA (%) |
| 20 | 0.65 | 93.32 ± 1.22 | 0.74 | 91.50 ± 2.01 | 0.99 | 89.62 ± 1.52 | 40 | 0.86 | 93.56 ± 1.00 | 0.83 | 92.06 ± 1.88 | 1.03 | 90.12 ± 0.94 | 60 | 1.17 | 95.34 ± 0.94 | 1.15 | 92.95 ± 1.05 | 1.64 | 90.95 ± 1.32 | 80 | 1.54 | 98.64 ± 1.26 | 1.44 | 93.67 ± 1.22 | 1.71 | 91.02 ± 0.98 | 100 | 1.32 | 98.96 ± 0.55 | 1.46 | 94.11 ± 1.14 | 1.92 | 92.45 ± 1.14 | 120 | 5.43 | 99.02 ± 1.25 | 4.98 | 94.21 ± 1.35 | 5.24 | 92.49 ± 1.32 |
|
|