Table 8.
Recognition accuracy (EER) of recognition systems using a combination of visible-light and thermal images for the recognition problem without PCA (unit: %).
| First SVM Layer Kernel | Accuracy of Recognition Systems Using Combined Visible-Light and Thermal Images | ||||||
|---|---|---|---|---|---|---|---|
| Feature-Level Fusion Approach | Score-Level Fusion Approach | ||||||
| EER | FAR | GAR | Second SVM layer kernel | Accuracy | |||
| EER | FAR | GAR | |||||
| Linear | 11.766 | 5.000 | 73.356 | Linear | 11.919 | 5.000 | 73.684 |
| 10.000 | 85.297 | ||||||
| 10.000 | 85.465 | 11.940 | 88.102 | ||||
| 15.000 | 91.214 | ||||||
| 11.780 | 88.247 | 20.000 | 94.388 | ||||
| RBF | 11.956 | 5.000 | 73.271 | ||||
| 15.000 | 91.280 | 10.000 | 85.339 | ||||
| 11.960 | 88.048 | ||||||
| 20.000 | 94.488 | 15.000 | 91.169 | ||||
| 20.000 | 94.618 | ||||||
| RBF | 11.684 | 5.000 | 73.394 | Linear | 11.850 | 5.000 | 74.008 |
| 10.000 | 85.206 | ||||||
| 10.000 | 85.689 | 11.860 | 88.160 | ||||
| 15.000 | 91.402 | ||||||
| 11.700 | 88.332 | 20.000 | 94.627 | ||||
| RBF | 11.956 | 5.000 | 73.201 | ||||
| 15.000 | 91.545 | 10.000 | 85.292 | ||||
| 11.960 | 88.049 | ||||||
| 20.000 | 94.692 | 15.000 | 91.288 | ||||
| 20.000 | 94.670 | ||||||