Scholarlink

<p>This paper presents the design and validation of an intelligent hardware–software system for automated pipe counting in continuous production. The solution employs a&#xa0;two-stage neural-network architecture based on the YOLOv8 model and implemented as a&#xa0;two-stage architecture that first localizes the region of interest (RoI) and then performs high-precision pipe detection. The system integrates robust image acquisition, adaptive processing algorithms, and real-time data handling to ensure accuracy and stability under industrial conditions. Both software and hardware components are described in detail. Field trials confirmed high performance, achieving a&#xa0;precision of 99.3%, recall of 99.7%, and an F1 score of 99.5%. The system contributes to the digitalization of quality control and improves the efficiency of pipe manufacturing operations.</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

Design of an intelligent hardware–software system for real-time pipe counting in continuous production

  • 期刊论文

  • 发表日期：

    2025年10月27日

查看全文
  • Igor Yu. Pyshmintsev,
  • Evgeniy A. Shkuratov,
  • Grigory A. Yashin,
  • Ilya V. Kos’min,
  • Sofiya K. Rosolenko,
  • Roman O. Bushin,
  • Vladimir L. Pyatkov,
  • Aleksandr V. Murzin,
  • Oleg P. Izgorev,
  • Evgeny V. Mazurin

摘要

This paper presents the design and validation of an intelligent hardware–software system for automated pipe counting in continuous production. The solution employs a two-stage neural-network architecture based on the YOLOv8 model and implemented as a two-stage architecture that first localizes the region of interest (RoI) and then performs high-precision pipe detection. The system integrates robust image acquisition, adaptive processing algorithms, and real-time data handling to ensure accuracy and stability under industrial conditions. Both software and hardware components are described in detail. Field trials confirmed high performance, achieving a precision of 99.3%, recall of 99.7%, and an F1 score of 99.5%. The system contributes to the digitalization of quality control and improves the efficiency of pipe manufacturing operations.