Table 10 Simulation results - data preprocessing impact on NASA dataset.
From: Real time fire and smoke detection using vision transformers and spatiotemporal learning
Model | Without preprocessing | With preprocessing |
|---|---|---|
Proposed hybrid model | 95.4% | 99.2% |
ResNet50 | 87.3% | 90.5% |
VGG16 | 83.5% | 87.6% |
LSTM | 89.3% | 91.3% |
3D-CNNs | 91.2% | 94.7% |
Hybrid ResNet50 + LSTM | 93.0% | 95.8% |
Hybrid VGG16 + 3D-CNN | 92.2% | 95.2% |
