Table 7. Classification accuracy (%) for all fused features using SVM, RF, and QDA classifiers for the BreakHis dataset.
Features | SVM | RF | QDA |
---|---|---|---|
Magnification Factor 40× | |||
Inception-ResNet-50 | 96.31 | 95.72 | 93.04 |
All fused features | 96.499 | 97.848 | 94.65 |
Magnification Factor 100× | |||
Inception-ResNet-V2 | 95.33 | 96.83 | 93.09 |
All fused features | 96.63 | 97.63 | 94.65 |
Magnification Factor 200× | |||
Inception-ResNet-V2 | 92.994 | 90.122 | 87.432 |
All fused features | 95.71 | 92.84 | 90.39 |
Magnification Factor 400× | |||
Inception-ResNet-V2 | 92.994 | 90.122 | 87.432 |
All fused features | 89.436 | 92.543 | 87.902 |