Mohamed Hassan

TextFinder

Semantic Segmentation

TextFinder Image Dataset

v4

2023-07-26 6:58pm

Generated on Jul 26, 2023

Dataset Split

Train Set 96%
768Images
Valid Set 4%
35Images
Test Set 0%
1Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640
Modify Classes: 1 remapped, 10 dropped

Augmentations

Outputs per training example: 2
90° Rotate: Clockwise, Counter-Clockwise
Crop: 0% Minimum Zoom, 13% Maximum Zoom
Shear: ±5° Horizontal, ±12° Vertical
Grayscale: Apply to 3% of images
Saturation: Between -25% and +25%
Brightness: Between -25% and +25%
Exposure: Between -10% and +10%
Blur: Up to 1px
Noise: Up to 2% of pixels