Table 3 Comparison results between traditional machine learning and CNNs
From: Automatic freezing-tolerant rapeseed material recognition using UAV images and deep learning
Type | Models | Accuracy (%) | Recall (%) | Precision (%) | F-score (%) | Testing times (ms) |
|---|---|---|---|---|---|---|
Traditional machine learning | SIFT+SVM | 67.72 | 30.85 | 19.59 | 23.97 | 2.66 |
SURF+SVM | 61.05 | 36.17 | 17.35 | 23.45 | 0.59 | |
SIFT+ANN | 76.32 | 15.96 | 21.43 | 18.2 9 | 7.84 | |
SURF+ANN | 72.98 | 12.77 | 18.46 | 15.09 | 4.41 | |
CNNs | AlexNet | 92.63 | 73.40 | 80.23 | 76.67 | 10.16 |
VGGNet16 | 93.16 | 73.40 | 83.13 | 77.97 | 10.88 | |
GoogLeNet | 92.45 | 67.02 | 84.00 | 74.56 | 22.32 | |
ResNet18 | 92.98 | 69.15 | 85.53 | 76.47 | 36.54 | |
ResNet50 | 93.33 | 79.79 | 79.79 | 79.79 | 47.25 |