Surface Defect Detection of Aluminum Profiles Based on Multiscale and Self-Attention Mechanisms
Sensors · 02.05.2024 · 24 (9) · 2914
DOI: 10.3390/s24092914 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, online erschienen am 02.05.2024. Verfasst von Yichuan Shao und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Intelligent Science and Engineering, Shenyang University, Shenyang 110044, China. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Basic Research Project of Liaoning Provincial Department of Education "Training and Application of Multimodal Deep Neural Network Models for Vertical Fields", Research on the Construction of a New Artificial Intelligence Technology and High Quality Edu-cation Service Supply System in the 14th Five Year Plan for Educa…, Basic Research Project of Liaoning Provincial Department of Education “Training and Application of Multimodal Deep Neural Network Models for Vertical Fields”.
Abstract (Originalsprache)
To address the various challenges in aluminum surface defect detection, such as multiscale intricacies, sensitivity to lighting variations, occlusion, and noise, this study proposes the AluDef-ClassNet model. Firstly, a Gaussian difference pyramid is utilized to capture multiscale image features. Secondly, a self-attention mechanism is introduced to enhance feature representation. Additionally, an improved residual network structure incorporating dilated convolutions is adopted to increase the receptive field, thereby enhancing the network’s ability to learn from extensive information. A small-scale dataset of high-quality aluminum surface defect images is acquired using a CCD camera. To better tackle the challenges in surface defect detection, advanced deep learning techniques and data augmentation strategies are employed. To address the difficulty of data labeling, a transfer learning approach based on fine-tuning is utilized, leveraging prior knowledge to enhance the efficiency and accuracy of model training. In dataset testing, our model achieved a classification accuracy of 97.6%, demonstrating significant advantages over other classification models.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Intelligent Science and Engineering, Shenyang University, Shenyang 110044, China.
- School of Information Engineering, Shenyang University, Shenyang 110044, China.
- School of Science, Shenyang University of Technology, Shenyang 110044, China.
ORCID-Profile: Le Zhang
Förderung
- Basic Research Project of Liaoning Provincial Department of Education "Training and Application of Multimodal Deep Neural Network Models for Vertical Fields"
- Research on the Construction of a New Artificial Intelligence Technology and High Quality Edu-cation Service Supply System in the 14th Five Year Plan for Educa…
- Basic Research Project of Liaoning Provincial Department of Education “Training and Application of Multimodal Deep Neural Network Models for Vertical Fields”
- Research on the Construction of a New Artificial Intelligence Technology and High Quality Education Service Supply System in the 14th Five Year Plan for Educat…
Zitieren
APA
Shao, Y., Fan, S., Zhao, Q., Zhang, L., & Sun, H. (2024). Surface Defect Detection of Aluminum Profiles Based on Multiscale and Self-Attention Mechanisms. Sensors, 24(9), 2914. https://doi.org/10.3390/s24092914
Vancouver
Shao Y, Fan S, Zhao Q, Zhang L, Sun H. Surface Defect Detection of Aluminum Profiles Based on Multiscale and Self-Attention Mechanisms. Sensors. 2024;24(9):2914. doi:10.3390/s24092914
BibTeX
@article{Shao2024,
author = {Shao, Yichuan and Fan, Shuo and Zhao, Qian and Zhang, Le and Sun, Haijing},
title = {Surface Defect Detection of Aluminum Profiles Based on Multiscale and Self-Attention Mechanisms},
journal = {Sensors},
year = {2024},
volume = {24},
number = {9},
pages = {2914},
doi = {10.3390/s24092914}
}
EUVIMED