Table 1 The data augmentation used in pretraining
From: Self-supervised pre-training for joint optic disc and cup segmentation via attention-aware network
Augmentation | Parameters |
|---|---|
Rotation | 0\(^{\circ }\), 90\(^{\circ }\), 180\(^{\circ }\), 270\(^{\circ }\) |
Sharpen | \(\alpha\) = [0, 1] |
GammaContrast | \(\gamma\) = [0.5, 2.0] |
GaussianBlur | \(\sigma\) = [0, 0.5] |