Skip to main content

Table 13 EER, FRR 1% , FRR 10% , AUC, HTER and Rank-1/5 identification rates of the quality-dependent approach using the different training sets for Protocol 2

From: MultiQ: single sensor-based multi-quality multi-modal large-scale biometric score database and its performance evaluation

Fusion rule Training set EER [%] FRR 1% [%] FRR 10% [%] AUC [%] HTER [%] Rank-1 [%] Rank-5 [%]
LLR (Q-stack) Set 1 13.0 69.8 18.9 5.5 12.5 31.1 48.1
  Set 2 13.0 73.9 19.3 5.7 14.6 29.1 45.8
  Set 3 13.8 53.7 18.2 5.9 15.7 40.2 55.9
  Set 4 15.8 50.6 21.8 7.3 19.2 41.7 56.7
  Set 5 13.1 69.2 18.9 5.5 12.6 32.2 49.3
  Set 6 13.3 74.3 19.7 5.8 13.5 30.0 47.0
  Set 7 14.1 52.2 18.6 6.1 16.6 41.2 56.9
  Set 8 16.2 49.3 22.4 7.7 20.3 42.3 57.5
  Set 9 12.9 69.3 18.8 5.4 12.4 29.5 46.3
  Set 10 13.0 73.6 19.0 5.7 14.0 27.5 43.6
  Set 11 13.4 53.6 17.7 5.6 15.0 39.3 55.1
  Set 12 15.2 50.9 20.8 6.9 18.0 41.1 56.2
  1. FRR 1% and FRR 10% are FRR at 1% FAR and FRR at 10% FAR respectively