random images Dataset

v3

2024-04-04 12:01pm

Generated on Apr 4, 2024

Dataset Split

Train Set 77%
18070Images
Valid Set 15%
3556Images
Test Set 7%
1704Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640
Auto-Adjust Contrast: Using Histogram Equalization
Tile: 2 rows x 2 columns
Filter Null: Require all images to contain annotations.

Augmentations

Outputs per training example: 2
Exposure: Between -15% and +15%
Noise: Up to 1.5% of pixels