Dataset Versions

v7

80-20_color_troceria

Generated on Sep 19, 2024

Dataset Split

Train Set 80%
3408Images
Valid Set 20%
859Images
Test Set %
0Images

Preprocessing

Auto-Orient: Applied
Static Crop: 0-100% Horizontal Region, 20-75% Vertical Region
Resize: Stretch to 1280x800
Auto-Adjust Contrast: Using Adaptive Equalization
Tile: 1 rows x 1 columns
Filter Null: Require all images to contain annotations.

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Rotation: Between -11° and +11°
Shear: ±15° Horizontal, ±15° Vertical
Brightness: Between -15% and +15%
Noise: Up to 2% of pixels