Welcome to the InfoSci Platform
Convolutional Neural Networks and Deep Learning Techniques for Glass Surface Defect Inspection

Convolutional Neural Networks and Deep Learning Techniques for Glass Surface Defect Inspection

Eduardo José Villegas-Jaramillo, Mauricio Orozco-Alzate
Copyright: © 2023 |Pages: 33
ISBN13: 9781668449912|ISBN10: 1668449919|EISBN13: 9781668449936
DOI: 10.4018/978-1-6684-4991-2.ch004
Cite Chapter Cite Chapter

MLA

Villegas-Jaramillo, Eduardo José, and Mauricio Orozco-Alzate. "Convolutional Neural Networks and Deep Learning Techniques for Glass Surface Defect Inspection." Revolutionizing Industrial Automation Through the Convergence of Artificial Intelligence and the Internet of Things, edited by Divya Upadhyay Mishra and Shanu Sharma, IGI Global, 2023, pp. 67-99. https://doi.org/10.4018/978-1-6684-4991-2.ch004

APA

Villegas-Jaramillo, E. J. & Orozco-Alzate, M. (2023). Convolutional Neural Networks and Deep Learning Techniques for Glass Surface Defect Inspection. In D. Mishra & S. Sharma (Eds.), Revolutionizing Industrial Automation Through the Convergence of Artificial Intelligence and the Internet of Things (pp. 67-99). IGI Global. https://doi.org/10.4018/978-1-6684-4991-2.ch004

Chicago

Villegas-Jaramillo, Eduardo José, and Mauricio Orozco-Alzate. "Convolutional Neural Networks and Deep Learning Techniques for Glass Surface Defect Inspection." In Revolutionizing Industrial Automation Through the Convergence of Artificial Intelligence and the Internet of Things, edited by Divya Upadhyay Mishra and Shanu Sharma, 67-99. Hershey, PA: IGI Global, 2023. https://doi.org/10.4018/978-1-6684-4991-2.ch004

Export Reference

Mendeley
Favorite

Abstract

Convolutional neural networks and their variants have revolutionized the field of image processing, allowing to find solutions to various types of problems in automatic visual inspection, such as, for instance, the detection and classification of surface defects in different types of industrial applications. In this chapter, a comparative study of different deep learning models aimed at solving the problem of classifying defects in images from a publicly available glass surface dataset is presented. Ten experiments were designed that allowed testing with several variants of the dataset, convolutional neural network architectures, residual learning-based networks, transfer learning, data augmentation, and (hyper)parameter tuning. The results show that the problem is difficult to solve due to both the nature of the defects and the ambiguity of the original class labels. All the experiments were analyzed in terms of different metrics for the sake of a better illustration and understanding of the compared alternatives.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.