china jiliang university

Badminton and Feather defect

Object Detection

Badminton and Feather defect Image Dataset

v2

defection and nodefection 1.1

Generated on May 25, 2023

Popular Download Formats

Dataset Split

Train Set 87%
1458Images
Valid Set 8%
138Images
Test Set 4%
75Images

Preprocessing

Resize: Stretch to 640x640
Grayscale: Applied
Filter Null: Require all images to contain annotations.

Augmentations

Outputs per training example: 3
Flip: Horizontal
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Crop: 0% Minimum Zoom, 16% Maximum Zoom
Rotation: Between -45° and +45°
Hue: Between -180° and +180°
Brightness: Between -33% and +33%
Exposure: Between -24% and +24%
Blur: Up to 3px
Noise: Up to 10% of pixels
Mosaic: Applied
Bounding Box: Flip: Horizontal, Vertical