Scholarlink

<p>Fast and accurate industrial Load Station (LS) inspection is crucial in manufacturing environments. This study addresses the challenges of deploying a deep learning-based solution for LS inspection in semiconductor wafer handling, where the Load Station must be properly aligned and occlusion-free before pin pack assembly operations. A significant challenge in industrial settings is data scarcity, as collecting and annotating large amounts of datasets is often impractical due to operational constraints and the rarity of abnormal conditions. This study examines how training data size affects inspection reliability in a real-world smart manufacturing context. The experiments utilise YOLOv5 and YOLOv8 variants across five training set sizes to determine minimum data requirements for reliable deployment, evaluated under 5-fold cross-validation with multiple random seeds. Our results demonstrate that YOLOv8 achieves superior data efficiency, with only 40 samples per class (<i>T</i>-40), YOLOv8n achieves <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(0.981 \pm 0.014\)</EquationSource> </InlineEquation> inspection accuracy and <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(0.842 \pm 0.050\)</EquationSource> </InlineEquation> mean Average Precision (mAP@0.5) on a held-out real test set, while YOLOv5 requires substantially more training data to achieve comparable performance. Smaller variants (YOLOv8n, YOLOv8s) consistently outperform larger models in this data-scarce environment. An ablation study confirms that combining real and synthetic obstruction samples is essential, and the approach is validated on an operational semiconductor manufacturing dataset, providing practical, statistically grounded recommendations for deploying deep learning inspection systems when training data are limited.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Deep learning based load station inspection for smart manufacturing with limited data

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月1日

查看全文
  • Yasir Ijaz,
  • Sonya Coleman,
  • Dermot Kerr,
  • Nazmul Siddique,
  • Cormac McAteer,
  • Bryan Baker,
  • Khoi Nguyen

摘要

Fast and accurate industrial Load Station (LS) inspection is crucial in manufacturing environments. This study addresses the challenges of deploying a deep learning-based solution for LS inspection in semiconductor wafer handling, where the Load Station must be properly aligned and occlusion-free before pin pack assembly operations. A significant challenge in industrial settings is data scarcity, as collecting and annotating large amounts of datasets is often impractical due to operational constraints and the rarity of abnormal conditions. This study examines how training data size affects inspection reliability in a real-world smart manufacturing context. The experiments utilise YOLOv5 and YOLOv8 variants across five training set sizes to determine minimum data requirements for reliable deployment, evaluated under 5-fold cross-validation with multiple random seeds. Our results demonstrate that YOLOv8 achieves superior data efficiency, with only 40 samples per class (T-40), YOLOv8n achieves \(0.981 \pm 0.014\) inspection accuracy and \(0.842 \pm 0.050\) mean Average Precision (mAP@0.5) on a held-out real test set, while YOLOv5 requires substantially more training data to achieve comparable performance. Smaller variants (YOLOv8n, YOLOv8s) consistently outperform larger models in this data-scarce environment. An ablation study confirms that combining real and synthetic obstruction samples is essential, and the approach is validated on an operational semiconductor manufacturing dataset, providing practical, statistically grounded recommendations for deploying deep learning inspection systems when training data are limited.