Journal of the Society of Naval Architects of Korea (대한조선학회논문집)

The Society of Naval Architects of Korea (대한조선학회)
권호 표지
DOI QR코드

DOI QR Code

Object Detection and Post-processing of LNGC CCS Scaffolding System using 3D Point Cloud Based on Deep Learning

딥러닝 기반 LNGC 화물창 스캐닝 점군 데이터의 비계 시스템 객체 탐지 및 후처리

  • Lee, Dong-Kun (Department of Naval Architecture and Ocean Engineering, Mokpo National Maritime University) ;
  • Ji, Seung-Hwan (Department of Ocean System Engineering, Graduate School, Mokpo National Maritime University) ;
  • Park, Bon-Yeong (Department of Ocean System Engineering, Graduate School, Mokpo National Maritime University)
  • 이동건 (목포해양대학교 조선해양공학과) ;
  • 지승환 (목포해양대학교 대학원 해양시스템공학과) ;
  • 박본영 (목포해양대학교 대학원 해양시스템공학과)
  • Received : 2021.05.24
  • Accepted : 2021.09.01
  • Published : 2021.10.20
Download PDF
⟨ Previous Next ⟩

Abstract

Recently, quality control of the Liquefied Natural Gas Carrier (LNGC) cargo hold and block-erection interference areas using 3D scanners have been performed, focusing on large shipyards and the international association of classification societies. In this study, as a part of the research on LNGC cargo hold quality management advancement, a study on deep-learning-based scaffolding system 3D point cloud object detection and post-processing were conducted using a LNGC cargo hold 3D point cloud. The scaffolding system point cloud object detection is based on the PointNet deep learning architecture that detects objects using point clouds, achieving 70% prediction accuracy. In addition, the possibility of improving the accuracy of object detection through parameter adjustment is confirmed, and the standard of Intersection over Union (IoU), an index for determining whether the object is the same, is achieved. To avoid the manual post-processing work, the object detection architecture allows automatic task performance and can achieve stable prediction accuracy through supplementation and improvement of learning data. In the future, an improved study will be conducted on not only the flat surface of the LNGC cargo hold but also complex systems such as curved surfaces, and the results are expected to be applicable in process progress automation rate monitoring and ship quality control.

Keywords

References

  1. Charles, R.Q. et al., 2018. Frustum PointNets for 3D object detection from rgb-d data. Proceeding of the IEEE Conference on Computer Vision and Pattern Recognition, arXiv: 1711.08488.
  2. Charles, R.Q. Hao, S. Kaichun, M. & Guibas, L.J., 2017. PointNet: Deep learning on point sets for 3D classification and segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, arXiv: 1612.00593.
  3. Ham, D. Lee, P. & Woo, J.H, 2016. A study of method for machine learning application for using shipyard big data. Journal of Computational Design and Engineering, pp.186-190.
  4. Jaderberg, M. Simonvan, K. Zisserman, A. & Kavukcuoglu, K., 2015. Spatial transformer networks. Part of Advances in Neural Information Processing Systems 28, paper ID:1213. Montreal, Canada, 7 - 12th December 2015.
  5. Jeong, Y.C. et al., 2021. Mechanical properties analysis of epoxy and polyurethane adhesive for accurate structural analysis of lng cargo hold. Journal of the Society of Naval Architects of Korea, 58(2), pp.66-72. https://doi.org/10.3744/SNAK.2021.58.2.066
  6. Kim, Y.J., Kim, T.W., Yoon, J.S. & Kim, M.K., 2019. Study of the construction of a coastal disaster prevention system using deep learning. Journal of Ocean Engineering and Technology, 33(6), pp.590-596. https://doi.org/10.26748/KSOE.2019.066
  7. Kirillov, A. et al., 2019. Panoptic segmentation. In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, arXiv: 1801.00868.
  8. Kwon, H. & Ruy, W., 2020. A study on the work-time estimation for block erections using stacking ensemble learning. Journal of the Society of Naval Architects of Korea, 56(6), pp.488-496. https://doi.org/10.3744/SNAK.2019.56.6.488
  9. Lee, D.K., Shin, J.G., Kim, Y. & Jeong, Y.K., 2014. Simulation-based work plan verification in shipyards. Journal of Ship Production and Design, 30(2), pp.49-57. https://doi.org/10.5957/JSPD.30.2.130032
  10. Maturana, D. & Scherer, S., 2015. VoxNet: A 3D convolution neural network for real-time object recognition. IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 922-928. Hamburg, Germany, 17 December 2015.
  11. Russakovsky, O. et al., 2015. ImageNet large scale visual recognition challenge. In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, arXiv: 1409.0575.
  12. Woo, J.H., Hwang, Y.S., & Nam, J.H., 2016. An approach for construction of shipyard simulation environment based on neutral file format. Journal of the Society of Naval Architects of Korea, 53(1), pp.18-28. https://doi.org/10.3744/SNAK.2016.53.1.18
  13. Yin, Z. & Tuzel, O., 2017. VoxelNet: End-to-end learning for point cloud based 3D object detection. Proceeding of the IEEE Conference on Computer Vision and Pattern Recognition, arXiv: 1711.06396.