Table 9 Fire in an industrial Building with limited visibility.
From: Real time fire and smoke detection using vision transformers and spatiotemporal learning
Model | Accuracy (%) | Precision (%) | Recall (%) | F1-Score (%) | AUC-ROC (%) | Specificity (%) |
|---|---|---|---|---|---|---|
Proposed hybrid model | 99.1 | 99.2 | 98.9 | 99.0 | 99.4 | 99.3 |
ResNet50 | 89.0 | 88.7 | 89.5 | 89.1 | 91.2 | 86.1 |
VGG16 | 87.2 | 85.4 | 88.1 | 86.7 | 90.0 | 85.4 |
LSTM | 91.1 | 91.2 | 90.6 | 90.9 | 92.4 | 88.3 |
3D-CNNs | 94.3 | 94.0 | 94.5 | 94.2 | 95.3 | 93.0 |
Hybrid ResNet50 + LSTM | 95.6 | 95.8 | 95.4 | 95.6 | 96.5 | 94.9 |
Hybrid VGG16 + 3D-CNN | 95.2 | 94.7 | 95.4 | 95.1 | 96.0 | 94.6 |
