드론을 활용한 실시간 원격 흘수 정밀 계측 프레임워크
(A Real-Time Draft Remote Measurement Framework using Unmanned Aerial Vehicles)
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
선박의 흘수 계측은 대량 화물 운송의 효율성을 높이고 흘수 깊이 초과로 인한 선박의 손상 등의 사고를 예방하는 중요한 작업이다. 기존 흘수 계측 방법은 작업자가 선박 외부에 근접하여 측정해야 하며, 이 과정에서 파도와 바람 등의 해양 환경으로 인해 사고가 발생할 위험이 있다. 이러한 문제를 해결하기 위해 본 논문에서는 드론을 활용한 자동 흘수 계측 프레임워크를 제안한다. 제안된 프레임워크는 드론의 위치를 추정하여 왜곡이 없는 흘수 정면 위치로 변환하고, segmentation foundation model을 활용해 훈련 없이 수면 선을 정확하게 검출한다. 이후 YOLO 모델을 통해 흘수를 탐지하고, 검출된 수면 선과 흘수를 바탕으로 최종 흘수 깊이를 계측한다. 실험 결과, 제안된 프레임워크는 드론 포즈 추정 및 흘수 계측 단계를 모두 포함하여 0.853초의 빠른 수행 시간을 보였다. 이는 기존 작업자가 수동으로 계측하는 방식에 비해 훨씬 짧은 시간으로, 실시간 적용이 가능함을 보여준다. 또한 평균 20mm 미만의 흘수 계측 오차는 현장 실무에서 적용될 수 있을 만큼 높은 정확도를 보여준다.
Abstract
Measuring a ship's draft is a critical task that enhances the efficiency of bulk cargo transportation and prevents accidents such as vessel damage due to exceeding maximum draft depth. Traditional draft measurement methods require workers to approach the exterior of the ship for measurement, posing risks of accidents caused by marine environmental factors like waves and wind. To address these issues, this paper proposes an automatic draft measurement framework utilizing drones. The proposed framework estimates the drone's position to transform the images into undistorted frontal views of the draft marks and accurately detects the waterline without the need for training by employing a segmentation foundation model. Subsequently, a YOLO model is used to detect the draft marks, and the final draft depth is measured based on the detected waterline and draft marks. Experimental results demonstrate that the proposed framework achieves a rapid execution time of 0.853 seconds, including both drone pose estimation and draft measurement stages. This is significantly shorter than the time required for manual measurement by workers, indicating its potential for real-time application. Moreover, an average draft measurement error of less than 20mm showcases a high level of accuracy suitable for practical field implementation.
참고문헌 (25)
- B. Wang, "Computer vision with deep learning for ship draft reading", Optical Engineering, 60(2), pp.024105~ (2021)
- J. Qu, "Multi-Task Learning-Enabled Automatic Vessel Draft Reading for Intelligent Maritime Surveillance", IEEE Transactions on Intelligent Transportation Systems, 25(5), pp.4552~4564 (2024)
- Y. Kirilenko, "Automatic Recognition of Draft Marks on a Ship’s Board Using Deep Learning System", International School on Neural Networks, 510, pp.1393~1401 (2022)
- R. Zang, "Searching Damaged Pine Trees by wilt Disease Based on Deep Learning Using Multispectral Image", Smart Media Journal, 45(11), pp.1823~1830 (2020)
- H. Myung, "Diagnosis of the Rice Loading for the UAV Image using Vision Transformer", Smart Media Journal, 12(9), pp.25~37 (2023)
- H. Kim, "Implementation of YOLO based Missing Person Search AI Application System", Smart Media Journal, 12(9), pp.159~170 (2023)
- K. He, "Mask R-CNN", Proceedings of the IEEE International Conference of Computer Vision, pp.2961~2969 (2017)
- O. Ronneberger, "U-net : Convolutional Networks for Biomedical Image Segmentation", International Conference Medical Image Computing and Computer-Assisted Intervention, pp.234~241 (2015)
- X. Liu, "Column-based Waterline Detection for Lightweight Ship Draft Reading", IEEE Transactions on Instrumentation and Measurement, 70, pp.1~10 (2022)
- B. Zhang, "Smart Ship Draft Reading by Dual-Flow Deep Learning Architecture Multispectral Information", Sensors, 24(17), pp.5580~ (2024)
- S. Wang, "DUSt3R : Geometric 3d Vision Made Easy", Proceedings of the IEEE, 2024, 20697~20709
- A. Kirillov. E, "Segment Anything", Proceedings of the IEEE, 2023, 4015~4026
- "YOLOv5"[URL]
- H. Zhang, "Evaluating the potential of post-processing kinematic(PPK)georeferencing for UAV-based structure-from-motion(SfM)photogrammetry and surface change detection", Earth Surface Dynamics Journal, 7(3), pp.807~827 (2019)
- C. Zhao, "Homography-based camera pose estimation with know gravity direction for UAV navigation", Science China Information Sciences, pp.1~13 (2021)
- R. Xiang, "Efficient Structure from Motion for Large-Size Videos from an Open Outdoor UAV Dataset", Sensors, (10) (2024)
- T. Tsujii, "Automatic Draft Reading based on Image Processing", Optical Engineering, 55(10), pp.104104~ (2016)
- J. L. H. Ferreira, "Character Recognition for Automatic Draught Survey", IEEE International Conference on Industry Applications (INDUSCON), 332(338), pp.332~338 (2018)
- X. Wang, "On Research of Video Stream Detection Algorithm for Ship Waterline", International Conference on Big Data, Artificial Intelligence and Internet of Things Engineering, pp.204~208 (2020)
- K. He, "Deep Residual Learning for Image Recognition", Proceedings of the IEEE, 2016, pp.770(778), 770~778
- M. Jeong, "Segmentation Foundation Model-baesd Automated Yard Management Algorithm", Smart Media Journal, 14(2), pp.52~61 (2024)
- "Number Dataset"[URL]
- P. E. Sarlin, "Pixel-Perfect Structure-From-Motion with Featuremetric Refinement", IEEE Transactions on Pattern Analysis and Machine Intelligence, pp, pp.1~12 (2023)