Journal of the Korean Society for Heat Treatment (열처리공학회지)

The Korean Society for Heat Treatment (한국열처리공학회)
권호 표지
DOI QR코드

DOI QR Code

Deep Learning-based Material Object Recognition Research for Steel Heat Treatment Parts

딥러닝 기반 객체 인식을 통한 철계 열처리 부품의 인지에 관한 연구

  • Hye-Jung, Park (Data Analysis and Research Team, P&S BigData Science Institute) ;
  • Chang-Ha, Hwang (Data Analysis and Research Team, P&S BigData Science Institute) ;
  • Sang-Gwon, Kim (Heat & Surface Technology R&D Department, Korea Institute of Industrial Technology (KITECH)) ;
  • Kuk-Hyun, Yeo (Heat & Surface Technology R&D Department, Korea Institute of Industrial Technology (KITECH)) ;
  • Sang-Woo, Seo (Data Analysis and Research Team, P&S BigData Science Institute)
  • 박혜정 (P&S빅데이터과학연구소) ;
  • 황창하 (P&S빅데이터과학연구소) ;
  • 김상권 (한국생산기술연구원 뿌리기술연구소) ;
  • 여국현 (한국생산기술연구원 뿌리기술연구소) ;
  • 서상우 (P&S빅데이터과학연구소)
  • Received : 2022.11.08
  • Accepted : 2022.11.21
  • Published : 2022.11.30
Download PDF
⟨ Previous Next ⟩

Abstract

In this study, a model for automatically recognizing several steel parts through a camera before charging materials was developed under the assumption that the temperature distribution in the pre-air atmosphere was known. For model development, datasets were collected in random environments and factories. In this study, the YOLO-v5 model, which is a YOLO model with strengths in real-time detection in the field of object detection, was used, and the disadvantages of taking a lot of time to collect images and learning models was solved through the transfer learning methods. The performance evaluation results of the derived model showed excellent performance of 0.927 based on mAP 0.5. The derived model will be applied to the model development study, which uses the model to accurately recognize the material and then match it with the temperature distribution in the atmosphere to determine whether the material layout is suitable before charging materials.

Keywords

Acknowledgement

본 논문은 한국생산기술연구원 기관주요사업 "Add-on 모듈 탑재를 통한 지능형 뿌리공정 기술개발 (KITECH EO-22-0005)"의 지원으로 수행한 연구입니다.

References

  1. Z. Zou, Z. Shi, Y. Guo, and J. Ye : Object detection in 20 years: A survey (2019).
  2. J. R. Uijlings, K. E. van de Sande, T. Gevers, and A. W. Smeulders : Selective search for object recognition, IJCV (2013).
  3. R. Girshick, J. Donahue, T. Darrell, and J. Malik : Rich feature hierarchies for accurate object detection and semantic segmentation, CVPR, (2014) 580-587.
  4. R. Girshick : Fast R-CNN, ICCV) (2015).
  5. S. Ren, K. He, R. Girshick, and J. Sun : Faster r-cnn: Towards real-time object detection with region proposal networks, NeurIPS (2015) 91-99.
  6. J. Redmon, S. Divvala, R. Girshick, and A. Farhadi : You only look once: Unified, real-time object detection, CVPR (2016) 779-788.
  7. J. Redmon and A. Farhadi : Yolo9000: Better, faster, stronger, CVPR (2017) 6517-6525.
  8. J. Redmon and A. Farhadi : Yolov3: An incremental improvement (2018) 1, 2, 3.
  9. G. Jocher et al. yolov5. https://github.com/ultralytics/yolov5 (2021) 1, 2, 3, 5, 6.