Comprehensive Quantitative Quality Assessment of Thermal Cut Sheet Edges using Convolutional Neural Networks
Janek Stahl (Fraunhofer IPA),* Andreas Frommknecht (Fraunhofer IPA), Marco Huber (University of Stuttgart)
The 34th British Machine Vision Conference
Abstract
In this study, we present a novel holistic approach to assess the quality of thermal cut edges using images of the cut edges. Using deep learning techniques, we estimate quality criteria such as roughness, edge slope tolerance, groove tracking, and burr height. Our approach significantly surpasses the current state of the art in evaluating thermal cut edges using 2D images. To the best of our knowledge, this study presents the first image-based groove tracking evaluation for thermal cut edges. Our results show that a comprehensive, accurate, and fast prediction of edge quality can be effectively achieved by implementing a simple image acquisition system combined with a convolutional neural network (CNN).Video
Citation
@inproceedings{Stahl_2023_BMVC,
author = {Janek Stahl and Andreas Frommknecht and Marco Huber},
title = {Comprehensive Quantitative Quality Assessment of Thermal Cut Sheet Edges using Convolutional Neural Networks},
booktitle = {34th British Machine Vision Conference 2023, {BMVC} 2023, Aberdeen, UK, November 20-24, 2023},
publisher = {BMVA},
year = {2023},
url = {https://papers.bmvc2023.org/0480.pdf}
}
Copyright © 2023
The British Machine Vision Conference is organised by