Research Article
A Hybrid Deep Learning Model for Trash Classification Based on Deep Trasnsfer Learning
Table 2
The test set accuracy of ResNeXt model and DNN-TC model.
| | Class | Classification accuracy of the ResNeXt-101 model | Classification accuracy of the DNN-TC model (%) |
| | Metal | 88.7% | 87.5% | | Pape | 85.2% | 83.4% | | Plastic | 81.3% | 79.4% | | Cardboard | 97.6% | 100% | | Glass | 94.4% | 95.4% | | Trash | 0 | 89.6% | | Total | 91.7% | 92.1% |
|
|
