Table 6 The feature extraction time (Fea_time) of different methods on the AR database injected with Gaussian noise
From: Noise-resistant network: a deep-learning method for face recognition under noise
| Â | FLBP | NRLBP | NRLBP+ | NRLBP++ | BN2 | BN1 | NR-Network | |
|---|---|---|---|---|---|---|---|---|
\( \sigma \) = 0.05 | Fea_time (s) | 0.6806 | 0.3988 | 0.7301 | 0. 7845 | 0.2694 | 0.2721 | 0.2743 |
\( \sigma \) = 0.10 | Fea_time (s) | 0.5990 | 0.2609 | 0.6365 | 0.6740 | – | – | – |
\( \sigma \) = 0.15 | Fea_time (s) | 0.5425 | 0.6832 | 1.2452 | 1.3580 | – | – | – |
\( \sigma \) = 0.20 | Fea_time (s) | 0.3553 | 0.2078 | 0.6405 | 0.6973 | – | – | – |