Deep Learning-Based Weld Contour and Defect Detection from Micrographs of Laser Beam Welded Semi-Finished Products

Laser beam welding is used in many areas of industry and research. There are many strategies and approaches to further improve the weld seam properties in laser beam welding. Metallography is often needed to evaluate welded seams. Typically, the images are examined and evaluated by experts. The eval...

Full description

Saved in:
Bibliographic Details
Published in:Applied sciences 2022-05, Vol.12 (9), p.4645
Main Authors: Nowroth, Christian, Gu, Tiansheng, Grajczak, Jan, Nothdurft, Sarah, Twiefel, Jens, Hermsdorf, Jörg, Kaierle, Stefan, Wallaschek, Jörg
Format: Article
Language:English
Subjects:
Artificial intelligence
Automatic welding
Contours
dataset creation
Datasets
Deep learning
Evaluation
Laser beam welding
Lasers
Measurement
Metallography
Micrography
Neural networks
Photomicrographs
quantization
Seams
Semantic segmentation
Software
weld defects
weld seam
Welded joints
Welding
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Laser beam welding is used in many areas of industry and research. There are many strategies and approaches to further improve the weld seam properties in laser beam welding. Metallography is often needed to evaluate welded seams. Typically, the images are examined and evaluated by experts. The evaluation process qualitatively provides the properties of the welds. Particularly in times when artificial intelligence is being used more and more in processes, the quantization of properties that could previously only be determined qualitatively is gaining importance. In this contribution, we propose to use deep learning to perform semantic segmentation of micrographs of complex weld areas to achieve the automatic detection and quantization of weld seam properties. A semantic segmentation dataset is created containing 282 labeled images. The training process is performed with DeepLabv3+. The trained model achieves a value of around 95% for weld contour detection and 76.88% of mean intersection over union (mIoU).
ISSN:2076-3417
2076-3417
DOI:10.3390/app12094645