Signature Extractor Dataset

v4

Signatures_with_stamps

Generated on Sep 7, 2024

Dataset Split

Train Set 88%
540Images
Valid Set 8%
46Images
Test Set 4%
25Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 256x256
Auto-Adjust Contrast: Using Adaptive Equalization
Grayscale: Applied
Filter Null: Require at least 10% of images to contain annotations.

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Crop: 0% Minimum Zoom, 18% Maximum Zoom
Rotation: Between -10° and +10°
Shear: ±9° Horizontal, ±13° Vertical
Exposure: Between -9% and +9%
Blur: Up to 0.7px
Noise: Up to 0.86% of pixels

Similar Projects

See More