Abstract
Surface defect detection is widely used in the manufacturing industry. Image-based surface defect classification has been proposed as a promising approach. Deep learning works well for various problems. However, for the problem of surface defect detection considered in this study, convolutional neural networks (CNNs) are not the best approach for all situations, and their performance can be further improved. The accurate defect detection provided by CNNs is high depending on the large dataset. When the dataset is small with limited samples, CNN performance cannot be guaranteed. Instead, a Vision Transformer model has been used in this paper. Experiments show that the Vision Transformer model performs better than CNN baselines (including VGG19, DenseNet, and ResNet) on an aluminum surface defect dataset.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
