Research Article
A Statistical Image Feature-Based Deep Belief Network for Fire Detection
Table 9
Precision, recall, and F1-score of the comparison study for fire detection.
| Pattern | AlexNet (%) | VGG16 (%) | VGG19 (%) | Precision | Recall | F1-score | Precision | Recall | F1-score | Precision | Recall | F1-score |
| Fire | 92.06 | 82.86 | 87.22 | 88.94 | 91.90 | 90.40 | 91.35 | 90.48 | 90.91 | Nonfire | 84.42 | 92.86 | 88.44 | 91.63 | 88.57 | 90.07 | 90.57 | 91.43 | 91.00 | Average | 88.24 | 87.86 | 87.83 | 90.28 | 90.24 | 90.24 | 90.96 | 90.95 | 90.95 |
|
|