Table 4 Performance analysis of different models for detecting an image.
From: A new method for Tomicus classification of forest pests based on improved ResNet50 algorithm
Model | Probability | Times (in seconds) |
|---|---|---|
ResNet18 | 81.3% | 0.1890 |
ResNet34 | 78.1% | 0.1502 |
ResNet50 | 81.7% | 0.1371 |
ResNet101 | 85.0% | 0.1456 |
MobileNetV2 | 82.0% | 0.1095 |
VGG16 | 90.5% | 0.3725 |
VGG19 | 86.2% | 0.3982 |
VIT-B/16 | 68.6% | 0.1749 |
EfficientNetV2 | 89.4% | 0.2466 |
AlexNet | 87.9% | 0.1568 |
DEMNet | 93.7% | 0.1193 |
