Table 5.
Comparison of finger-vein recognition accuracy (unit: %) (The “number*” is referred from [19]).
| Method | EER | ||
|---|---|---|---|
| SDU-DB | PolyU-DB | ||
| Non-training based method | Maximum Curvature [13] | 4.54* | 3.51 |
| Repeated line tracking [15] | 5.46* | 2.17 | |
| Wide line detector [16] | 22.7* | 1.80 | |
| Gabor + LBP [18] | 8.096 | 3.61 | |
| Training-based method | VGG Net-16 [32] | 3.906 | 2.491 |
| ResNet-50 | 3.4931 | 1.3435 | |
| ResNet-101 | 3.3653 | 1.0779 | |
| Score-level fusion of ResNet-50 and ResNet-101 by weighted product rule | 3.0653 | 0.8888 | |
| Non-training and training-based method | Score-level fusion of Gabor + LBP [18] and ResNet-101 by weighted product rule | 3.2426 | 0.9138 |