image classification dataset on tailored textiles quality control

Zenodo (CERN European Organization for Nuclear Research)

Abstract

This dataset was geared towards representing practical quality control scenarios, specifically involving the quality inspection of glass fiber fabric. Continuous rolls of glass fiber fabric were cut into samples of 300x200 mm. Half of these samples were reinforced with a single carbon fiber. These samples were then classified into six different categories based on the presence of common defects or if they were error-free textiles. Each category consists of 300 images, with a resolution of 4288x2848 pixels.

Authors 6

1. Andrés Felipe Posada-Moreno Aachen Institute for Data Science in Mechanical Engineering

   RWTH Aachen University

   Affiliation as printed

   Institute for Data Science in Mechanical Engineering, RWTH Aachen University

2. Kai Müller Aachen Institut für Textiltechnik

   RWTH Aachen University

   Affiliation as printed

   Institut für Textiltechnik of RWTH Aachen University

3. Florian Brillowski Aachen Institut für Textiltechnik

   RWTH Aachen University

   Affiliation as printed

   Institut für Textiltechnik of RWTH Aachen University

4. Friedrich Solowjow Aachen Institute for Data Science in Mechanical Engineering

   RWTH Aachen University

   Affiliation as printed

   Institute for Data Science in Mechanical Engineering, RWTH Aachen University

5. Thomas Gries Aachen Institut für Textiltechnik

   RWTH Aachen University

   Affiliation as printed

   Institut für Textiltechnik of RWTH Aachen University

6. Sebastian Trimpe Aachen Institute for Data Science in Mechanical Engineering

   RWTH Aachen University

   Affiliation as printed

   Institute for Data Science in Mechanical Engineering, RWTH Aachen University