Automated Quality Control of Vacuum Insulated Glazing by Convolutional Neural Network Image Classification

• Published in: arXiv.org 2021-10
• Main Authors: Riedel, Henrik, Mokdad, Sleheddine, Schulz, Isabell, Kocer, Cenk, Rosendahl, Philipp, Schneider, Jens, Kraus, Michael A, Drass, Michael
• Format: Article
• Language: English
• Subjects: Artificial neural networks; Atmospheric models; Automatic control; Automation; Computer vision; Damage assessment; Damage detection; Explainable artificial intelligence; Image classification; Machine learning; Neural networks; Quality control; Service life; Vacuum glazing; Vision systems
• ISSN: 2331-8422
• DOI: 10.48550/arxiv.2110.08079

Vacuum Insulated Glazing (VIG) is a highly thermally insulating window technology, which boasts an extremely thin profile and lower weight as compared to gas-filled insulated glazing units of equivalent performance. The VIG is a double-pane configuration with a submillimeter vacuum gap between the panes and therefore under constant atmospheric pressure over their service life. Small pillars are positioned between the panes to maintain the gap, which can damage the glass reducing the lifetime of the VIG unit. To efficiently assess any surface damage on the glass, an automated damage detection system is highly desirable. For the purpose of classifying the damage, we have developed, trained, and tested a deep learning computer vision system using convolutional neural networks. The classification model flawlessly classified the test dataset with an area under the curve (AUC) for the receiver operating characteristic (ROC) of 100%. We have automatically cropped the images down to their relevant information by using Faster-RCNN to locate the position of the pillars. We employ the state-of-the-art methods Grad-CAM and Score-CAM of explainable Artificial Intelligence (XAI) to provide an understanding of the internal mechanisms and were able to show that our classifier outperforms ResNet50V2 for identification of crack locations and geometry. The proposed methods can therefore be used to detect systematic defects even without large amounts of training data. Further analyses of our model's predictive capabilities demonstrates its superiority over state-of-the-art models (ResNet50V2, ResNet101V2 and ResNet152V2) in terms of convergence speed, accuracy, precision at 100% recall and AUC for ROC.