Skip to main content
zenodoopen

Underwater Plastic dataset

<p>The dataset was generated using the <a href="https://roboflow.com/">Roboflow</a> platform. The annotations are compatible with the <a href="https://github.com/ultralytics/yolov5">PyTorch YOLOv5</a> architecture.</p> <p><em><strong>Dataset details:</strong></em></p> <p><strong># Images:</strong> 1220 images</p> <p><strong>Image Split:</strong></p> <p>Train / Test Split: 92</p> <p>Training Set: 1.1k</p> <p><strong>Preprocessing</strong></p> <p>Auto-Orient: Applied</p> <p>Resize: Stretch to 416x416</p> <p>Augmentations</p> <p>Outputs per training example: 5</p> <p>Flip: Horizontal, Vertical</p> <p>Crop: 0% Minimum Zoom, 49% Maximum Zoom</p> <p>Grayscale: Apply to 47% of images</p> <p>Hue: Between -25&deg; and +25&deg;</p> <p>Saturation: Between -42% and +42%</p> <p>Exposure: Between -22% and +22%</p> <p>Blur: Up to 3.25px</p> <p>Cutout: 8 boxes with 10% size each</p> <p>Mosaic: Applied</p> <p>Details</p> <p>Version Name: 2022-07-24 12:50am</p> <p>Version ID: 1</p> <p>Generated: Jul 24, 2022</p>

ShareScore

40/100

Overall dataset sharing score

Score breakdown

These five areas show where the dataset supports — or may limit — practical reuse.

Stewardship
8
Harmonization
4
Access
16
Reuse readiness
8
Engagement
4

Topics

Dataset details

Repository
zenodo
Published
July 25, 2022
License
cc-by-4.0
DOI
10.5281/zenodo.6907229

Scored with release 0.9.0

ScienceDex for faculty

Your academic CV and NIH biosketch

Make your CV in your school's format and your NIH biosketch from it, then download both as Word files.