Mask R-CNN Weld Defect 3 Dataset

v1

2024-05-20 3:43pm

Generated on May 20, 2024

Dataset Split

Train Set 92%
192Images
Valid Set 6%
12Images
Test Set 2%
4Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640
Auto-Adjust Contrast: Using Histogram Equalization
Tile: 2 rows x 2 columns

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise
Crop: 0% Minimum Zoom, 25% Maximum Zoom
Rotation: Between -10° and +10°
Shear: ±10° Horizontal, ±10° Vertical
Grayscale: Apply to 15% of images
Brightness: Between -20% and +20%
Exposure: Between -10% and +10%
Blur: Up to 2.5px
Noise: Up to 1.01% of pixels

Similar Projects

See More