Speroid segmentation Computer Vision Dataset
99 Total Images
View All ImagesDataset Split
Train Set 88%
87Images
Valid Set 8%
8Images
Test Set 4%
4Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 512x512
Grayscale: Applied
Auto-Adjust Contrast: Using Adaptive Equalization
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Rotation: Between -20° and +20°
Blur: Up to 2px
Noise: Up to 1.96% of pixels