KIPS Transactions on Computer and Communication Systems (정보처리학회논문지:컴퓨터 및 통신 시스템)

Korea Information Processing Society (한국정보처리학회)
권호 표지
DOI QR코드

DOI QR Code

Shipping Container Load State and Accident Risk Detection Techniques Based Deep Learning

딥러닝 기반 컨테이너 적재 정렬 상태 및 사고 위험도 검출 기법

  • Received : 2022.09.02
  • Accepted : 2022.09.26
  • Published : 2022.11.30
Download PDF
⟨ Previous Next ⟩

Abstract

Incorrectly loaded containers can easily knock down by strong winds. Container collapse accidents can lead to material damage and paralysis of the port system. In this paper, We propose a deep learning-based container loading state and accident risk detection technique. Using Darknet-based YOLO, the container load status identifies in real-time through corner casting on the top and bottom of the container, and the risk of accidents notifies the manager. We present criteria for classifying container alignment states and select efficient learning algorithms based on inference speed, classification accuracy, detection accuracy, and FPS in real embedded devices in the same environment. The study found that YOLOv4 had a weaker inference speed and performance of FPS than YOLOv3, but showed strong performance in classification accuracy and detection accuracy.

최근 항만에서는 부정확한 컨테이너 적재로 인해 컨테이너가 강풍에 쉽게 쓰러지는 컨테이너 붕괴 사고가 빈번이 발생하고 있으며 이는 물적 피해와 항만 시스템 마비로 이어지고 있다. 본 논문에서는 이런 사고를 미연에 방지하기 위해 딥러닝 기반 컨테이너 적재 상태 및 사고 위험도 검출 시스템을 제안한다. 제안된 시스템은 darknet 기반 YOLO 모델을 활용하여 컨테이너 상하의 코너캐스팅을 통해 컨테이너 정렬 상태를 실시간으로 파악하고 관리자에게 사고 위험도를 알리는 시스템이다. 제안된 시스템은 추론 속도, 분류 정확도, 검출 정확도 등을 성능 지표와 실제 구현 환경에서 최적의 성능을 보인 YOLOv4 모델을 객체 인식 알고리즘 모델로 선택하였다. 제안된 알고리즘인 YOLOv4가 YOLOv3보다 추론속도와 FPS의 성능 측면에서 낮은 성능을 보이기는 했지만, 분류 정확도와 검출 정확도에서 강력한 성능을 보임을 증명하였다.

Keywords

Acknowledgement

이 논문은 연구개발특구 진흥재단의 지역연구개발혁신지원 지능형 무인 자동화 스마트물류 시스템 구축 과제(2020-DD-UP-0281-03-210)와 2022년도 BB21+ 사업으로 지원되었음.

References

  1. T. H. Lee, "Smart port policy trend of europe and singapore and its political implications," Journal of Korea Port Economic Association, Vol.36, No.1, pp.77-90, 2020. https://doi.org/10.38121/kpea.2020.03.36.1.77
  2. C. A. Duran, C. Fernandez-Campusano, R. Carrasco, M. Vargas, and A. Navarrete, "Boosting the decision-making in smart ports by using blockchain," IEEE Access, Vol.9, pp.128055-128068, 2021. https://doi.org/10.1109/ACCESS.2021.3112899
  3. E. Nishimura, "Yard arrangement problem with the external truck arrival," 2021 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM), pp.88-92, 2021.
  4. G. Parise, L. Parise, L. Martirano, P. B. Chavdarian, C. -L. Su, and A. Ferrante, "Wise port and business energy management: Port facilities, electrical power distribution," IEEE Transactions on Industry Applications, Vol.52, No.1, pp.18-24, 2016. https://doi.org/10.1109/TIA.2015.2461176
  5. K. -L. A. Yau, S. Peng, J. Qadir, Y. -C. Low, and M. H. Ling, "Towards smart port infrastructures: enhancing port activities using information and communications technology," IEEE Access, Vol.8, pp.83387-83404, 2020. https://doi.org/10.1109/access.2020.2990961
  6. Container Cargo Processing Performance (2022) [Internet], https://www.index.go.kr/potal/main/EachDtlPageDetail.d o?idx_cd=1267#quick_02.
  7. S. Y. Oh, J. H. Jeong, B. Choi, J. H. Yeon, Y. U. Seo, S. W. Kim, and J. S. Youn, "A study on deep learning model for container load status monitoring," Proceedings of the Annual Spring Conference of Korea Information Processing Society Conference (KIPS) 2022, Vol.29, pp.320-321, 2022.
  8. J. Zhao, S. Hao, C. Dai, H. Zhang, L. Zhao, Z. Ji, and I. Ganchev, "Improved vision-based vehicle detection and classification by optimized YOLOv4," IEEE Access, Vol.10, pp.8590-8603, 2022. https://doi.org/10.1109/ACCESS.2022.3143365
  9. J. Fan, J. Lee, I. Jung, and Y. Lee, "Improvement of object detection based on faster R-CNN and YOLO," 2021 36th International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC), pp.1-4, 2021.
  10. S. Ren, K. He, R. Girshick, and J. Sun, "Faster R-CNN: Towards real-time object detection with region proposal networks," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol.39, No.6, pp.1137-1149, 2017. https://doi.org/10.1109/TPAMI.2016.2577031
  11. Q. Guo, L. Liu, W. Xu, Y. Gong, X. Zhang, and W. Jing, "An improved faster R-CNN for high-speed railway dropper detection," IEEE Access, Vol.8, pp.105622-105633, 2020. https://doi.org/10.1109/access.2020.3000506
  12. Y. H. Lee and Y. S. Kim, "Comparison of CNN and YOLO for Object Detection," Journal of the Semiconductor & Display Technology, Vol.19, No.1, pp.85-92, 2020.
  13. Y. Cai et al., "YOLOv4-5D: An Effective and Efficient Object Detector for Autonomous Driving," IEEE Transactions on Instrumentation and Measurement, Vol.70, pp.1-13, 2021.
  14. J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, "You only look once: Unified, real-time object detection," In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.779-788, 2016.
  15. J. Redmon and A. Farhadi, "YOLO9000: Better, faster, stronger," In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.7263-7271, 2017.
  16. J. Redmon and A. Farhadi, Yolov3: An incremental improvement, arXiv preprint arXiv:1804.02767, 2018.
  17. Z. Cong and X. Li, "Track obstacle detection algorithm based on YOLOv3," 2020 13th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), pp.12-17, 2020.
  18. H. Wang, H. Pei, and J. Zhang, "Detection of locomotive signal lights and pedestrians on railway tracks using improved YOLOv4," IEEE Access, Vol.10, pp.15495-15505, 2022. https://doi.org/10.1109/ACCESS.2022.3148182
  19. T. Yang and C. Tong, "Small Traffic Sign Detector in Real-time Based on Improved YOLO-v4," 2021 IEEE 23rd Int Conf on High Performance Computing & Communications; 7th Int Conf on Data Science & Systems; 19th Int Conf on Smart City; 7th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys), pp.1318-1324, 2021.
  20. A. Bochkovskiy, C. Y. Wang, and H. Y. M. Liao, Yolov4: Optimal speed and accuracy of object detection, arXiv preprint arXiv:2004.10934, 2020.