Automated optical inspection based on synthetic mechanisms combining deep learning and machine learning

Chung-Ming Lo () and Ting-Yi Lin
Additional contact information
Chung-Ming Lo: National Chengchi University
Ting-Yi Lin: National Chengchi University

Journal of Intelligent Manufacturing, 2025, vol. 36, issue 7, No 15, 4769-4783

Abstract: Abstract The quality inspection of products before delivery plays a critical role in ensuring manufacturing quality. Quick and accurate inspection of samples is realized by highly automated inspection based on pattern recognition in smart manufacturing. Conventional ensemble methods have been demonstrated to be effective for defect detection. This study further proposed synthetic mechanisms based on using various features and learning classifiers. A database of 6000 sample images of printed circuit board (PCB) connectors collected from factories was compiled. A novel confidence synthesis mechanism was proposed to prescreen images using deep learning features. Spatially connected texture features were then used to reclassify images with low reliabilities. The synthetic mechanism was found to outperform a single classifier. In particular, the highest improvement in accuracy (from 96.00 to 97.83%) was obtained using the confidence-based synthesis. The synthetic mechanism can be used to achieve high accuracy in defect detection and make automation in smart manufacturing more practicable.

Keywords: Machine learning; Deep learning; AOI; Synthetic mechanisms (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s10845-024-02474-4 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:joinma:v:36:y:2025:i:7:d:10.1007_s10845-024-02474-4

Ordering information: This journal article can be ordered from
http://www.springer.com/journal/10845

DOI: 10.1007/s10845-024-02474-4

Access Statistics for this article

Journal of Intelligent Manufacturing is currently edited by Andrew Kusiak

More articles in Journal of Intelligent Manufacturing from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().