CNN 기반의 선저 표면 손상 유형 분류 모델

양민우¹ ; 김두영²^{, *}

1해군사관학교 인공지능학과 사관생도
2해군 중령(진)/해군사관학교 인공지능학과 부교수

CNN-based Classification Model of Ship Hull Surface Damage

Minwoo Yang¹ ; Dooyoung Kim²^{, *}

1Midshipman, Dept. of Artificial Intelligence, Republic of Korea Naval Academy
2CDR(Sel.), ROK Navy/Associate professor, Dept. of Artificial Intelligence, Republic of Korea Naval Academy

Correspondence to: ^*Dooyoung Kim Dept. of Artificial Intelligence, Republic of Korea Naval Academy 1, Jungwon-ro, Jinhae-gu, Changwon-si, Gyungsangnam-do, 51704, Republic of Korea Tel: +82-2-907-5246 E-mail: dykim07@navy.ac.kr

초록

본 논문에서는 GRN 기반의 CNN 모델을 통해 함정 표면의 손상을 분류하고, 그 분류 근거를 시각화하여 제시하였다. 먼저, 자체 CNN 모델을 설계하여 선박의 손상 6개의 클래스를 분류하였다. 실험 결과, 정확도 97.6 %로 6개의 클래스를 분류하였다. 다음으로 자체 CNN 모델의 분류 근거를 Grad-CAM을 활용해 시각화하였다.

Abstract

This study presented a GRN-based customized CNN model to classify naval/maritime ship damage and to visualize the decision basis of the classification. The custom CNN classified ship damage into six categories, achieving a classification accuracy of 97.6 %. Furthermore, the decision-making process of the model was visualized using Grad-CAM, which highlighted the key regions contributing to the classification.

Keywords:

Convolutional Neural Network, Explainable AI, Hull Inspection, Classification Problem, Lightweight Deep Learning

키워드:

합성곱 신경망, 설명 가능한 AI, 선저 검사, 분류 문제, 경량 딥러닝

References

국토교통부, 선박검사제도 개요. (2012). https://www.molit.go.kr/USR/policyData/m_34681/dtl?id=143&
선박안전법 시행규칙 제19조 2항. (2025).
선박안전법 제60조. (2023).
조원진, “선박 검사 프로세스에 AI 도입⋯디지털 전환 전략 속대 낸다,” 서울경제 (2023. 12. 14). https://sedaily.com/NewsView/29YIDECXCC
Widiarti, E., Fitriyani, D., Rachmawati, R.(2024). “Corrosion Detection on Ship Hull Using ROV Based on Convolutional Neural Network,” International Journal of Marine Engineering Innovation and Research, Vol. 9, No. 1, pp. 1-7. [https://doi.org/10.12962/j25481479.v9i1.17235]
Li, X., Zhang, Y., Wang, H., Chen, J. (2024). “A Zero-Shot Image Classification Method of Ship Coating Defects Based on IDATLWGAN,” Journal of Marine Science and Engineering, Vol. 12, No. 3, 456.
Chin, C. S., Si, J., Clare, A. S., Ma, M. (2017). “Intelligent Image Recognition System for Marine Biofouling Using Softmax Transfer Learning and Deep Convolutional Neural Networks,” Complexity, 2017, Article ID 5730419. [https://doi.org/10.1155/2017/5730419]
International Maritime Organization. (2023). “Guidelines for the Control and Management of Ships’s Biofouling to Minimize the Transfer of Invasive Aquatic Species (MEPC.378(80)).”
American Bureau of Shipping. (2017). “Guidance Notes on the Inspection, Maintenance and Application of Marine Coating Systems. American Bureau of Shipping.”
Hull integrity and Ship Maintenance Software – ShipManager Hull. https://dnv.com/services/hull-integrity-and-ship-maintenance-software-shipmanager-hull-1531/
International Association of Classification Societies. 2014. “Recommendation No. 42 Rev.1 – Guidelines for Surveys, Assessment and Repair of Hull Structure.”
DNV. (2016). “Rules for Classification: Ships, Part 3 Chapter 1 – General Principles.”
Automating the Assessment of Biofouling in Images and Video Footage [Dataset]. (2024). https://figshare.com/articles/dataset/Automating_the_assessment_of_biofouling_in_images/26537158
LIACi Semantic Segmentation Dataset for Underwater Ship Inspections. (2024). https://data.sintef.no/product/dp-9e112cec-3a59-4b58-86b3-ecb1f2878c60
MaVeCoDD Dataset: Marine Vessel Hull Corrosion in Dry-Dock Images. (2021). https://data.mendeley.com/datasets/ry392rp8cj/1
Fushi Computer Vision Project. (2025). https://universe.roboflow.com/liangs/fushi-045qu
선박 도장 품질 데이터. (2022). https://www.aihub.or.kr/aihubdata/data/view.do?currMenu=115&topMenu=100&dataSetSn=71447
Ioffe, S., & Szegedy, C. (2015). “Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate shift,” In Proceedings of the 32nd International Conference on Machine Learning (ICML). arXiv: 1502.03167
Krizhevsky, A. (2009). “Convolutional Deep Belief Networks on CIFAR-10.” Technical Report.
Yamaguchi, T., Takahashi, S., Minematsu, N., & Hirose, K. (1990). “Neural Network for Speaker-Independent Isolated Word Recognition,” IEEE Transactions on Acoustics, Speech, and Signal Processing, Vol. 38, No. 11, pp. 1842-1845.
Chollet, F. (2017). “Xception: Deep Learning with Depthwise Separable Convolutions,” In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.1251-1258. [https://doi.org/10.1109/CVPR.2017.195]
Xu, Y., et al. (2014). “Scale-Invariant Convolutional Neural Networks.” arXiv:1411.6369, .
Woo, S., et al. (2024). “ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders.” arXiv preprint:2405.03301v1, . [https://doi.org/10.1109/CVPR52729.2023.01548]
Lee, C.-Y., Gallagher P.W., & Tu, Z. (2015). “Generalizing Pooling Functions in Convolutional Neural Networks: Mixed, Gated, and Tree,” In Proceedings of the 18th International Conference on Artificial Intelligence and Statistics (AISTATS).
Yue, S., et al. (2022). “Imbalanced Malware Images Classification: a CNN Based Approach,” arXiv:1708.08042, .
Kingma D. P., & Ba, J. (2014). Adam: A method for Stochastic Optimization.