Dataset Versions

v3

2024-10-29 12:07am

Generated on Oct 29, 2024

Dataset Split

Train Set 97%
3735Images
Valid Set 2%
81Images
Test Set 1%
40Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640
Auto-Adjust Contrast: Using Histogram Equalization

Augmentations

Outputs per training example: 3
90° Rotate: Clockwise, Counter-Clockwise
Saturation: Between -12% and +12%
Brightness: Between -23% and +23%
Exposure: Between -15% and +15%