Abstract
Ship classification based on AIS trajectory data is an important aspect of spatio-temporal trajectory data mining. Aiming at the fact that most of the features extracted by the existing ship classification methods are motion features, which ignore the spatial relation between the vessels and the coastline, a Method based on LightGBM (Light Gradient Boosting Machine) for ship classification considering the offshore distance features is proposed. First, the trajectory data is cleaned and segmented, then the basic motion features of the trajectory segment are extracted, and the features are further enhanced by combining the offshore distance of the vessels. After being normalized, the features are filtered by chi-square test method, and finally a LightGBM model is constructed. The selected features are used for classification training of five types of vessels: cargo, passenger, fishing, tug and yacht, and stable classification results are obtained by 5-fold cross-validation. The results show that the average accuracy of ship classification based on LightGBM model is 86.6% after the offshore distance feature is added, and the classification accuracy is improved by 3.1%, which is better than the traditional machine classification methods such as Random Forest, Decision Tree and KNN model.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Chen, R.L., Wang, Y.Q., Liu, B.J., et al.: Spatio-temporal features of vessel activities in the Bohai Sea based on GIS and AIS. Progr. Geogr. 39(07), 1172–1181 (2020)
Zhen, R., Shao, Z.P., Pan, J.C.: Advance in character mining and prediction of vessel behavior based on AIS data. J. Geoinf. Sci. 23(12), 2111–2127 (2021)
Zheng, Q.L., Fan, W., Zhang, S.M., et al.: Identification of fishing type from VMS data based on artificial neural network. South China Fish. Sci. 12(02), 81–87 (2016)
Sánchez, P.D., Amigo, D., GarcÃa, J., et al.: Architecture for trajectory-based fishing vessel classification with AIS data. Sensors 20(13), 3782–3802 (2020)
Ljunggren, H.: Using deep learning for classifying vessel trajectories. In: 21st International Conference on Information Fusion, pp. 2158–2164. IEEE, Cambridge (2018)
Bakkegaard, S., Blixenkrone, M.J., Larsen, J.J., et al.: Target classification using kinematic data and a recurrent neural network. In: 19th International Radar Symposium (IRS), pp. 20–22. IEEE, Bonn (2018)
Yung, P.C., Irvine, J.M.: Passive identification of vessel type through track motion analysis. In: Geospatial Informatics X, pp. 20–40. SPIE, Washington (2020)
Feng, C., Fu, B., Luo, Y., et al.: The design and development of a vessel trajectory data management and analysis system based on AIS. Sensors 22(1), 310–331 (2021)
Sheng, K., Liu, Z., Zhou, D., et al.: Research on vessel classification based on trajectory features. J. Navig. 71(1), 100–116 (2018)
Kraus, P., Mohrdieck, C., Schwenker, F.: Ship classification based on trajectory data with machine-learning methods. In: 19th International Radar Symposium (IRS), pp. 1–10. IEEE, Bonn (2018)
Ginoulhac, R., Barbaresco, F., Schneider, J., et al.: Coastal radar target recognition based on kinematic data (AIS) with machine learning. In: 2019 International Radar Conference (RADAR), pp. 1–5. IEEE, Bonn (2019)
Yang, T., Wang, X., Liu, Z.: Ship type recognition based on vessel navigating trajectory and convolutional neural network. J. Mar. Sci. Eng. 10(1), 84–104 (2022)
Li, T., Xu, H., Zeng, W.: Ship classification method for massive AIS trajectories based on GNN. J. Phys: Conf. Ser. 2025(1), 12–24 (2021)
Wang, Y., Yang, L., Song, X.: Ship classification for space-based AIS data using 1D-CNN. In: 5th International Conference on Electronic Information Technology and Computer Engineering, pp. 840–844. ACM, Xiamen (2021)
Duan, H., Ma, F., Miao, L., et al.: A semi-supervised deep learning approach for vessel trajectory classification based on AIS data. Ocean Coast. Manag. 218(3), 1–12 (2022)
Chang, J.L., Xie, L., Wei, Z.W., et al.: Research on vessel trajectory classification based on deep convolution neural network. J. Wuhan Univ. Technol. (Transp. Sci. Eng.) 1(46), 160–165 (2022)
Cui, T.T., Wang, G.L., Gao, J.: Vessel trajectory classification method based on 1DCNN-LSTM. Comput. Sci. 47(9), 175–184 (2020)
Damastuti, N., Aisjah, A.S., Masroeri, A.: Vessel classifying and trajectory based on automatic identification system data. IOP Conf. Ser.: Earth Environ. Sci. 830(1), 12–49 (2021)
Zhong, H., Song, X., Yang, L.: Vessel classification from Space-based AIS data using random forest. In: 5th International Conference on Big Data and Information Analytics (BigDIA), pp. 9–12. IEEE, Kunming (2019)
Lin, J.Y., Zheng, B.L., Liu, J.: Ship classification model based on convolutional neural network. Inf. Technol. Inform. (2), 125–126 (2019)
Friedman, J.H.: Greedy function approximation: boosting machine. Annals 2001 29(5), 1189–1232 (2001)
Wang, H., Jiang, Y.N., Zhang, X., et al.: Lithology identification method based on gradient boosting algorithm. J. Jilin Univ. (Earth Sci. Ed.) 51(3), 940–950 (2021)
Yu, Q., Huang, X.L.: Classification of heart sound signals based on LightGBM. J. Shaanxi Normal Univ. (Nat. Sci. Ed.) 48(6), 47–55 (2020)
Singh, M., Nicholas, E., Singh, M., et al.: Using spectral acoustic features to identify abnormal heart sounds. In: 2016 Computing in Cardiology Conference (CinC), pp. 557–560. IEEE, Vancouver (2016)
Fang, M., Meng, X.F.: Data Mining Concepts and Technologies. China Machine Press, Beijing (2019)
Sang, L.Z., Wall, A., Mao, Z., et al.: A novel method for restoring the trajectory of the inland waterway vessel by using AIS data. Ocean Eng. 110(1), 183–194 (2015)
Li, H.: Statistical Learning Methods, 2nd edn. Tsinghua University Press, Beijing (2019)
Acknowledgements
This project is supported by National Natural Science Foundation of China (41901397, 42101454, 42101455), and supported by Institute of Geospatial Information, University of Information Engineering.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Xu, L., Chen, X., Wen, B., Ma, J., Wang, Y., Xu, Q. (2023). Ship Classification Based on Trajectories Data and LightGBM Considering Offshore Distance Feature. In: Meng, X., et al. Spatial Data and Intelligence. SpatialDI 2023. Lecture Notes in Computer Science, vol 13887. Springer, Cham. https://doi.org/10.1007/978-3-031-32910-4_8
Download citation
DOI: https://doi.org/10.1007/978-3-031-32910-4_8
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-32909-8
Online ISBN: 978-3-031-32910-4
eBook Packages: Computer ScienceComputer Science (R0)