target_detect Dataset

v3

2023-05-17 7:19pm

Generated on May 17, 2023

Dataset Split

Train Set 87%
3984Images
Valid Set 8%
376Images
Test Set 5%
216Images

Preprocessing

Auto-Orient: Applied
Resize: Fit within 640x640
Auto-Adjust Contrast: Using Contrast Stretching
Grayscale: Applied
Tile: 2 rows x 2 columns

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Crop: 0% Minimum Zoom, 20% Maximum Zoom
Rotation: Between -15° and +15°
Shear: ±15° Horizontal, ±15° Vertical
Brightness: Between -25% and +25%
Blur: Up to 1px
Noise: Up to 3% of pixels
Cutout: 10 boxes with 5% size each