Research Article
Hyperspectral Image Classification Model Using Squeeze and Excitation Network with Deep Learning
Table 7
Evaluation of accuracy on each class of Salinas dataset with overall accuracy.
| Class | VGG-16 | Inception-v3 | ResNet-50 | Proposed |
| Brocoli_green_weeds_1 | 98.74 | 99.56 | 99.38 | 99.65 | Brocoli_green_weeds_2 | 97.49 | 99.89 | 99.60 | 99.96 | Fallow | 98.50 | 99.93 | 100 | 99.92 | Fallow_rough_plow | 98.87 | 99.50 | 99.10 | 99.44 | Fallow_smooth | 98.79 | 98.27 | 99.74 | 99.50 | Stubble | 96.00 | 96.80 | 99.28 | 99.81 | Celery | 98.94 | 99.13 | 99.41 | 99.90 | Grapes_untrained | 86.52 | 91.95 | 93.82 | 94.14 | Soil_vinyard_develop | 99.87 | 100 | 100 | 100 | Corn_senesced_green_weeds | 94.38 | 97.58 | 98.25 | 97.40 | Lettuce_romaine_4 wk | 93.82 | 97.75 | 99.01 | 98.57 | Lettuce_romaine_5 wk | 97.06 | 99.49 | 99.72 | 99.80 | Lettuce_romaine_6 wk | 98.31 | 99.24 | 98.90 | 99.35 | Lettuce_romaine_7 wk | 91.95 | 98.89 | 99.05 | 99.26 | Vinyard_untrained | 65.15 | 78.36 | 81.29 | 82.90 | Vinyard_vertical_trellis | 97.87 | 98.21 | 99.00 | 99.28 | Overall accuracy | 90.43 | 92.89 | 95.15 | 96.33 |
|
|