Fig 1.
Structure of original GhostNet model.
a. The convolutional layer, b. The Ghost module.
Fig 2.
Structure of Ghost bottlenecks.
Table 1.
Original overall architecture of GhostNet.
Table 2.
Modified overall architecture of GhostNet.
Table 3.
The parameters of UAV images dataset.
Fig 3.
Combination of rotation and noise interference.
a) 90° + high noise, b) 180° + high noise, c) 270° + high noise, d) 90° + low noise, e) 180° + low noise, f) 270° + low noise.
Fig 4.
Combination of rotation and different brightness and contrast augmentation.
(a) 90° + α = 1.4, β = 0.6. (b) 180° + α = 1.4, β = 0.6. (c) 270° + α = 1.4, β = 0.6. (d) 90° + α = 0.8, β = 1.2. (e) 180° + α = 0.8, β = 1.2. (f) 270° + α = 0.8, β = 1.2.
Fig 5.
Feature maps derived from different layers.
(a) The 8th layer. (b) The 9th layer. (c) The 10th layer. (d) The 11th layer.
Fig 6.
The entire training strategy of the Modified GhostNet model.
Table 4.
The samples of a single category in training set before and after augmentation.
Table 5.
The average accuracy of GhostNet before and after image augmentation.
Fig 7.
The loss and accuracy of the GhostNet model before and after UCMerced image augmentation.
(a) Loss of training set. (b) Accuracy of training set. (c) Loss of validation set. (d) Accuracy of validation set.
Fig 8.
The loss and accuracy of the GhostNet model before and after AID image augmentation.
(a) Loss of training set. (b) Accuracy of training set. (c) Loss of validation set. (d) Accuracy of validation set.
Fig 9.
The loss and accuracy of the GhostNet model before and after NWPU-RESISC image augmentation.
(a) Loss of training set. (b) Accuracy of training set. (c) Loss of validation set. (d) Accuracy of validation set.
Table 6.
FLOPs of different models on UCMerced dataset.
Table 7.
The memory usage and predicted time of different models.
Table 8.
The average accuracy of different models.
Fig 10.
The average accuracy of dropout position.
Fig 11.
The average accuracy of the model before and after using transfer learning.