| Titre : |
Deep learning-based automated characterization of crosscut tests for coatings via image segmentation |
| Type de document : |
texte imprimé |
| Auteurs : |
Gaoyuan Zhang, Auteur ; Christian Schmitz, Auteur ; Matthias Fimmers, Auteur ; Christoph Quix, Auteur ; Sayed Hoseini, Auteur |
| Année de publication : |
2022 |
| Article en page(s) : |
p. 671-683 |
| Note générale : |
Bibliogr. |
| Langues : |
Américain (ame) |
| Catégories : |
Adhésion
Analyse d'imageL'analyse d'image est la reconnaissance des éléments contenus dans l'image. Il ne faut pas confondre analyse (décomposition en éléments) et traitement (action sur les composantes) de l'image.
Analyse de dommages (matériaux)
Autoapprentissage
Automatisation
Caractérisation
Enrobage (technologie)
Essais de quadrillage
Imagerie (technique)
Logiciels
Rayures
Résistance à l'abrasion
Revêtements -- Détérioration
Surfaces (Physique)
|
| Index. décimale : |
667.9 Revêtements et enduits |
| Résumé : |
A manual scratch test to measure the scratch resistance of coatings applied to a certain substrate is usually used to test the adhesion of a coating. Despite its significant amount of subjectivity, the crosscut test is widely considered to be the most practical measuring method for adhesion strength with a good reliability. Intelligent software tools help to improve and optimize systems combining chemistry, engineering based on high-throughput formulation screening (HTFS) technologies and machine learning algorithms to open up novel solutions in material sciences. Nevertheless, automated testing often misses the link to quality control by the human eye that is sensitive in spotting and evaluating defects as it is the case in the crosscut test. In this paper, we present a method for the automated and objective characterization of coatings to drive and support Chemistry 4.0 solutions via semantic image segmentation using deep convolutional networks. The algorithm evaluated the adhesion strength based on the images of the crosscuts recognizing the delaminated area and the results were compared with the traditional classification rated by the human expert. |
| Note de contenu : |
- BACKGROUND : Neural networks - Convolutional neural networks and image segmentation tasks
- SOLUTION APPROACH AND EXPERIMENTAL SETUP : Sample preparation - Automated crosscutting - Data preprocessing - Data augmentation - Architectures Loss functions
- EVALUATION : Training - Results
- Table 1 : Summary of the applied data augmentation methods
- Table 2 : Confusion matrices of the various models being applied for the segmentation into delaminated and intact area
- Table 3 : Comparison between human and algorithmic rating for samples selected across all six levels defined in the norm
- Table 4 : Mean Dice coefficient in % calculated over all 31 test images with standard deviation, inference time for the test set and the number of parameters for each model paired with each of the different losses |
| DOI : |
https://doi.org/10.1007/s11998-021-00557-y |
| En ligne : |
https://link.springer.com/content/pdf/10.1007/s11998-021-00557-y.pdf |
| Format de la ressource électronique : |
Pdf |
| Permalink : |
https://e-campus.itech.fr/pmb/opac_css/index.php?lvl=notice_display&id=37296 |
in JOURNAL OF COATINGS TECHNOLOGY AND RESEARCH > Vol. 19, N° 2 (03/2022) . - p. 671-683
Aller sur edunet
Accueil
Deep learning-based automated characterization of crosscut tests for coatings via image segmentation in JOURNAL OF COATINGS TECHNOLOGY AND RESEARCH, Vol. 19, N° 2 (03/2022)
