Abrasive, adhesive, and fatigue wear datasets for Artificial Neural Network training

Published: 10 August 2026| Version 1 | DOI: 10.17632/598ptz86sh.1
Contributors:
,

Description

COMLGAN (Concurrent Optimization with Mutual Learning GAN) was trained using datasets of abrasive, adhesive, and fatigue wear. From the original SEM photos of friction/wear tests, all images exhibiting particular wear damages were manually chosen and cropped.

Files

Institutions

Categories

Delamination, Abrasive Wear, Adhesive Wear, Fatigue Crack, Deep Learning, Generative Adversarial Network

Licence