A Performance Comparison of State-of-the-Art Models in Ship Detection  from Aerial Images

Quang Dung Bui; Kinh Bac Dang; Manh Ha Nguyen; Thi Huyen Ngoc Hoang; Giang Tuan Linh; Van Trong Giang; Mohamad Yasir; Phuc Tuan Tong; Cong Long Nguyen; Thu Nhung Nguyen; Thai Son Nguyen; Khanh Toan Tran; Thi Huong Giang Le

doi:10.15625/1859-3097/22402

Author affiliations

Authors

Quang Dung Bui Institute of Earth Sciences, VAST, Vietnam; Graduate University of Science and Technology, VAST, Vietnam
Kinh Bac Dang University of Science, Vietnam National University, Hanoi, Vietnam
Manh Ha Nguyen Institute of Earth Sciences, VAST, Vietnam; Graduate University of Science and Technology, VAST, Vietnam
Thi Huyen Ngoc Hoang Institute of Earth Sciences, VAST, Vietnam; Graduate University of Science and Technology, VAST, Vietnam
Tuan Linh Giang University of Science, Vietnam National University, Hanoi, Vietnam; VNU Institute of Vietnamese Studies and Development Sciences, Vietnam National University, Hanoi, Vietnam
Van Trong Giang VNU Institute of Vietnamese Studies and Development Sciences, Vietnam National University, Hanoi, Vietnam
Mohamad Yasir College of Oceanography and Space Informatics, China University of Petroleum, Qingdao 266580, China
Phuc Tuan Tong Institute of Earth Sciences, VAST, Vietnam
Cong Long Nguyen Institute of Earth Sciences, VAST, Vietnam
Thu Nhung Nguyen Institute of Earth Sciences, VAST, Vietnam
Thai Son Nguyen Institute of Earth Sciences, VAST, Vietnam
Khanh Toan Tran Vietnam Maritime University, Hai Phong, Vietnam
Thi Huong Giang Le Vietnam Maritime University, Hai Phong, Vietnam

DOI:

https://doi.org/10.15625/1859-3097/22402

Keywords:

Ship, vessel, monitor, object detection, deep learning, Vietnam

Abstract

In the face of rapidly advancing technology, environmental concerns, and societal changes, the maritime sector faces many challenges in providing safe, environmentally friendly shipping, fishing, and offshore services. The dynamic nature of fishing zones, the limited technology aboard some vessels, the lack of data in developing regions, and the prevalence of illicit activities all contribute to the difficulty of ship detection in maritime environments. This research presents an optimized deep learning model to tackle these issues head-on, thereby improving the ship detection system’s precision and effectiveness in Vietnam using high-resolution images acquired by the Worldview-2 sensor. The research seeks to answer questions about real-time vessel and fishing-area detection and to investigate the future distribution of fishing areas. Three network architectures, including Fast R-CNN, YOLOv9, and SSD, generate twelve models. After optimization, the model with a Fast R-CNN architecture, 128 × 128 input size, and no non-maximum suppression (NMS) was used to detect ships and estimate their density in central Vietnam, achieving 96% accuracy. YOLOv9 (128 × 128) should be used for real-time vessel detection despite the model’s accuracy of approximately 92%, which can be crucial for managing fisheries activities in other regions.

Downloads

Download data is not yet available.

References

[1] R. W. Liu, W. Yuan, X. Chen, and Y. Lu, “An enhanced CNN-enabled learning method for promoting ship detection in maritime surveillance system,” Ocean Engineering, vol. 235, 109435, 2021.

[2] A. Böhnke-Henrichs, C. Baulcomb, R. Koss, S. S. Hussain, and R. S. de Groot, “Typology and indicators of ecosystem services for marine spatial planning and management,” Journal of Environmental Management, vol. 130, pp. 135–145, 2013.

[3] F. Gabel, S. Lorenz, and S. Stoll, “Effects of ship-induced waves on aquatic ecosystems,” Science of the Total Environment, vol. 601,

pp. 926–939, 2017.

[4] Q. Jiang and H. Li, “Silicon energy bulk material cargo ship detection and tracking method combining YOLOv5 and DeepSort,” Energy Reports, vol. 9, pp. 151–158, 2023.

[5] T. M. Gutiérrez, P. J. Castillo, B. L. Naranjo, and M. J. Akester, “Current state of goods, services and governance of the Humboldt Current Large Marine Ecosystem in the context of climate change,” Environmental Development, vol. 22, pp. 175–190, 2017.

[6] Q. T. K. Ngoc, “Impacts on the ecosystem and human well-being of the marine protected area in Cu Lao Cham, Vietnam,” Marine Policy, vol. 90, pp. 174–183, 2018.

[7] Federal Geographic Data Committee, Coastal and Marine Ecological Classification Standard. Reston, VA, USA, 2012. [Online]. Available: https://www.fgdc.gov/standards/projects/cmecs-folder/CMECS_Version_06-2012_FINAL.pdf

[8] A. A. Kurekin, B. R. Loveday, O. Clements, G. D. Quartly, P. I. Miller, G. Wiafe, and K. Adu Agyekum, “Operational monitoring of illegal fishing in Ghana through exploitation of satellite earth observation and AIS data,” Remote Sensing, vol. 11, no. 3, 293, 2019.

[9] M. D. Ferreira, G. Spadon, A. Soares, and S. Matwin, “A semi-supervised methodology for fishing activity detection using the geometry behind the trajectory of multiple vessels,” Sensors, vol. 22, no. 16, 6063, 2022.

[10] M. Coll, L. J. Shannon, K. M. Kleisner, M. J. Juan-Jordá, A. Bundy, A. G. Akoglu, D. Banaru, J. L. Boldt, M. F. Borges, A. Cook, I. Diallo, C. Fu, C. Fox, D. Gascuel, L. J. Gurney, T. Hattab, J. J. Heymans, D. Jouffre, B. R. Knight, S. Kucukavsar, S. I. Large, C. Lynam, A. Machias, K. N. Marshall, H. Masski, H. Ojaveer, C. Piroddi, J. Tam, D. Thiao, M. Thiaw, M. A. Torres, M. Travers-Trolet, K. Tsagarakis, I. Tuck, G. I. van der Meeren, D. Yemane, S. G. Zador, and Y. J. Shin, “Ecological indicators to capture the effects of fishing on biodiversity and conservation status of marine ecosystems,” Ecological Indicators, vol. 60, pp. 947–962, 2016.

[11] L. T. Nguyen, P. D. Tran, and K. Q. Nguyen, “An effectiveness of artificial coral reefs in the restoration of marine living resources,” Regional Studies in Marine Science, vol. 49, 102143, 2022.

[12] Z. Chen, D. Chen, Y. Zhang, X. Cheng, M. Zhang, and C. Wu, “Deep learning for autonomous ship-oriented small ship detection,” Safety Science, vol. 130, 104812, 2020.

[13] G. Matasci, J. Plante, K. Kasa, P. Mousavi, A. Stewart, A. Macdonald, A. Webster, and J. Busler, “Deep learning for vessel detection and identification from spaceborne optical imagery,” ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. 3, pp. 303–310, 2021.

[14] A. J. Gallego, A. Pertusa, and P. Gil, “Automatic ship classification from optical aerial images with convolutional neural networks,” Remote Sensing, vol. 10, no. 4, 511, 2018.

[15] T. Zhang, X. Zhang, C. Liu, J. Shi, S. Wei, I. Ahmad, X. Zhan, Y. Zhou, D. Pan, J. Li, and H. Su, “Balance learning for ship detection from synthetic aperture radar remote sensing imagery,” ISPRS Journal of Photogrammetry and Remote Sensing, vol. 182, pp. 190–207, 2021.

[16] Q. Shi, W. Li, R. Tao, X. Sun, and L. Gao, “Ship classification based on multifeature ensemble with convolutional neural network,” Remote Sensing, vol. 11, no. 4, 419, 2019.

[17] W. Li, Z. Wang, and H. Huang, “Seasonal and intra-seasonal variations of surface hydrological parameters over North Yellow Sea and its implications for applicability of quasi-synchronous ship-based multi-station hydrological investigation data,” Journal of Sea Research, vol. 190, 102310, 2022.

[18] P. N. Pathak, “Comparison of sea-surface temperature observations from TIROS-N and ships in the North Indian Ocean during MONEX (May–July 1979),” Remote Sensing of Environment, vol. 12, no. 5, pp. 363–369, 1982.

[19] J. A. Bowker and N. C. Townsend, “Evaluation of bow foils on ship delivered power in waves using model tests,” Applied Ocean Research, vol. 123, 103148, 2022.

[20] E. Ruth and I. Thompson, “Comparing design assumptions with hindcast wave conditions and measured ship speed and heading,” Ocean Engineering, vol. 247, 110613, 2022.

[21] I. Grappasonni, F. Petrelli, and F. Amenta, “Deaths on board ships assisted by the Centro Internazionale Radio Medico in the last 25 years,” Travel Medicine and Infectious Disease, vol. 10, no. 4, pp. 186–191, 2012.

[22] J. Escorcia-Gutierrez, M. Gamarra, K. Beleño, C. Soto, and R. F. Mansour, “Intelligent deep learning-enabled autonomous small ship detection and classification model,” Computers and Electrical Engineering, vol. 100, 107871, 2022.

[23] M. Yasir, L. Shanwei, X. Mingming, S. Hui, M. S. Hossain, A. T. I. Colak, D. Wang, W. Jianhua, and K. B. Dang, “Multi-scale ship target detection using SAR images based on improved YOLOv5,” Frontiers in Marine Science, vol. 9, 1086140, 2023.

[24] A. Rawson, M. Brito, Z. Sabeur, and L. Tran-Thanh, “A machine learning approach for monitoring ship safety in extreme weather events,” Safety Science, vol. 141, 105336, 2021.

[25] P. Thoya, B. Kaunda-Arara, J. Omukoto, C. Munga, E. Kimani, and A. O. Tuda, “Trawling effort distribution and influence of vessel monitoring system (VMS) in Malindi-Ungwana Bay: Implications for resource management and marine spatial planning in Kenya,” Marine Policy, vol. 109, 103677, 2019.

[26] S. E. Birchenough, P. A. Cooper, and A. C. Jensen, “Vessel monitoring systems as a tool for mapping fishing effort for a small inshore fishery operating within a marine protected area,” Marine Policy, vol. 124, 104325, 2021.

[27] M. Bellanger and H. Levrel, “A cost-effectiveness analysis of alternative survey methods used for the monitoring of marine recreational fishing in France,” Ocean & Coastal Management, vol. 138, pp. 19–28, 2017.

[28] M. D. M. Galotto-Tébar, A. Pomares-Padilla, I. A. Czerwinski, and J. C. Gutiérrez-Estrada, “Is the vessel fishing? Discrimination of fishing activity with low-cost intelligent mobile devices through traditional and heuristic approaches,” Expert Systems with Applications, vol. 200, 117091, 2022.

[29] X. Lou, Y. Liu, Z. Xiong, and H. Wang, “Generative knowledge transfer for ship detection in SAR images,” Computers and Electrical Engineering, vol. 101, 108041, 2022.

[30] T. P. Lai, “Fisheries co-management in Vietnam: Issues and approach,” in Proceedings of the International Institute of Fisheries Economics and Trade (IIFET) Conference, Vietnam Policy Day, Nha Trang, Vietnam, 2008.

[31] M. Ibrion, N. Paltrinieri, and A. R. Nejad, “Learning from failures in cruise ship industry: The blackout of Viking Sky in Hustadvika, Norway,” Engineering Failure Analysis, vol. 125, 105355, 2021.

[32] J. Aminuddin, R. F. Abdullatif, E. I. Anggraini, S. F. Gumelar, and A. Rahmawati, “Development of convolutional neural network algorithm on ships detection in Natuna Islands-Indonesia using land look satellite imagery,” Remote Sensing Applications: Society and Environment, vol. 32, 101025, 2023.

[33] F. Zeng, L. Cheng, N. Li, N. Xia, L. Ma, X. Zhou, and M. Li, “A hierarchical airport detection method using spatial analysis and deep learning,” Remote Sensing, vol. 11, no. 19, 2204, 2019.

[34] N. Hieu, D. Van Bao, V. Van Vinh, T. P. N. Pham, D. T. Phuong, and P. H. T. Thi, “Geomorphological resources for tourism development in central Highlands and South Central coast,” Vietnam Journal of Earth Sciences, vol. 1S, pp. 35–47, 2015. [in Vietnamese].

[35] V. P. Vu, H. Nguyen, T. V. Hoang, B. Nguyen, and M. T. Dao, “Some results of geomorphological research on the modern coast of Vietnam,” VNU Journal of Science: Earth and Environmental Sciences, vol. 4, pp. 73–81, 2004. [in Vietnamese].

[36] H. H. T. Thu, B. D. Kinh, and A. Van Rompaey, “Comprehensive assessment of coastal tourism potential in Vietnam,” Vietnam Journal of Earth Sciences, vol. 44, no. 4, pp. 535–558, 2022.

[37] L. Hens, N. A. Thinh, T. H. Hanh, N. S. Cuong, T. D. Lan, N. Van Thanh, and D. T. Le, “Sea-level rise and resilience in Vietnam and the Asia-Pacific: A synthesis,” Vietnam Journal of Earth Sciences, vol. 40, no. 2, pp. 126–152, 2018.

[38] P. Van Hung, “Potential effects of the Kra Canal on Vietnam’s maritime industry,” Transportation Research Interdisciplinary Perspectives, vol. 14, 100622, 2022.

[39] B. B. Xuan, Q. T. K. Ngoc, and T. Börger, “Fisher preferences for marine litter interventions in Vietnam,” Ecological Economics, vol. 200, 107534, 2022.

[40] S. Albawi, T. A. Mohammed, and S. Al-Zawi, “Understanding of a convolutional neural network,” in Proc. 2017 International Conference on Engineering and Technology (ICET), 2018, pp. 1–6.

[41] T. L. Giang, Q. T. Bui, T. D. L. Nguyen, V. B. Dang, Q. H. Truong, T. T. Phan, H. Nguyen, V. L. Ngo, V. T. Tran, M. Yasir, and K. B. Dang, “Coastal landscape classification using convolutional neural network and remote sensing data in Vietnam,” Journal of Environmental Management, vol. 335, 117537, 2023. DOI: 10.1016/j.jenvman.2023.117537.

[42] K. B. Dang, Q. T. Bui, T. P. N. Pham, and V. L. Ngo, “A convolutional neural network for coastal classification based on ALOS and NOAA satellite data,” IEEE Access, vol. 8, pp. 11824–11839, 2020.

[43] H. Perez, J. H. Tah, and A. Mosavi, “Deep learning for detecting building defects using convolutional neural networks,” Sensors,

vol. 19, no. 16, 3556, 2019.

[44] T. Ming and Y. Ju, “SAR ship detection based on YOLOv5,” in Proc. Third International Conference on Computer Vision and Data Mining (ICCVDM 2022), vol. 12511. Bellingham, WA, USA: SPIE, 2023, pp. 408–413.

[45] Z. Liu, R. Feng, L. Wang, Y. Zhong, and L. Cao, “D-ResUNet: ResUNet and dilated convolution for high resolution satellite imagery road extraction,” in Proc. 2019 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), 2019, pp. 3927–3930.

[46] D. Biswas, H. Su, C. Wang, A. Stevanovic, and W. Wang, “An automatic traffic density estimation using Single Shot Detection (SSD) and MobileNet-SSD,” Physics and Chemistry of the Earth, Parts A/B/C, vol. 110, pp. 176–184, 2019. DOI: 10.1016/j.pce.2018.12.001.

[47] F. Zeng, Y. Liu, Y. Ye, J. Zhou, and X. Liu, “A detection method of Edge Coherent Mode based on improved SSD,” Fusion Engineering and Design, vol. 179, p. 113141, 2022.

[48] Q. Zhang, X. Hu, Y. Yue, Y. Gu, and Y. Sun, “Multi-object detection at night for traffic investigations based on improved SSD framework,” Heliyon, vol. 8, no. 11, e11570, 2022.

[49] K. B. Dang, K. C. Vu, H. Nguyen, D. A. Nguyen, T. D. L. Nguyen, T. P. N. Pham, T. L. Giang, H. D. Nguyen, and T. H. Do, “Application of deep learning models to detect coastlines and shorelines,” Journal of Environmental Management, vol. 320, 115732, 2022.

[50] H. Guo, X. Yang, N. Wang, and X. Gao, “A CenterNet++ model for ship detection in SAR images,” Pattern Recognition, vol. 112, 107787, 2021.