Research on Ship Classification Based on Trajectory Features

It is important for maritime authorities to effectively classify and identify unknown types of ships in historical trajectory data. This paper uses a logistic regression model to construct a ship classifier by utilising the features extracted from ship trajectories. First of all, three basic movement patterns are proposed according to ship sailing characteristics, with related sub-trajectory partitioning algorithms. Subsequently, three categories of trajectory features with their extraction methods are presented. Finally, a case study on building a model for classifying fishing boats and cargo ships based on real Automatic Identification System (AIS) data is given. Experimental results indicate that the proposed classification method can meet the needs of recognising uncertain types of targets in historical trajectory data, laying a foundation for further research on camouflaged ship identification, behaviour pattern mining, outlier behaviour detection and other applications.

[1]  Maarten van Someren,et al.  Machine learning for vessel trajectories using compression, alignments and domain knowledge , 2012, Expert Syst. Appl..

[2]  Yu Zheng,et al.  Trajectory Data Mining , 2015, ACM Trans. Intell. Syst. Technol..

[3]  Xing Xie,et al.  Inferring Taxi Status Using GPS Trajectories , 2012, ArXiv.

[4]  Yanmin Zhu,et al.  A Survey on Trajectory Data Mining: Techniques and Applications , 2016, IEEE Access.

[5]  Jae-Gil Lee,et al.  Trajectory clustering: a partition-and-group framework , 2007, SIGMOD '07.

[6]  Nicholas Jing Yuan,et al.  T-Finder: A Recommender System for Finding Passengers and Vacant Taxis , 2013, IEEE Transactions on Knowledge and Data Engineering.

[7]  Robert Weibel,et al.  Revealing the physics of movement: Comparing the similarity of movement characteristics of different types of moving objects , 2009, Comput. Environ. Urban Syst..

[8]  Torgeir Moan,et al.  Estimating Navigation Patterns from AIS , 2009, Journal of Navigation.

[9]  Brian Sullivan,et al.  Ending hide and seek at sea , 2016, Science.

[10]  Jae-Gil Lee,et al.  TraClass: trajectory classification using hierarchical region-based and trajectory-based clustering , 2008, Proc. VLDB Endow..

[11]  Trevor Hastie,et al.  The Elements of Statistical Learning , 2001 .

[12]  Lyle H. Ungar,et al.  Machine Learning manuscript No. (will be inserted by the editor) Active Learning for Logistic Regression: , 2007 .

[13]  Xuelong Li,et al.  When Location Meets Social Multimedia , 2015, ACM Transactions on Intelligent Systems and Technology.

[14]  P. Silveira,et al.  Use of AIS Data to Characterise Marine Traffic Patterns and Ship Collision Risk off the Coast of Portugal , 2013, Journal of Navigation.

[15]  Maarten van Someren,et al.  An analysis of alignment and integral based kernels for machine learning from vessel trajectories , 2014, Expert Syst. Appl..

[16]  Mong-Li Lee,et al.  Incorporating Duration Information for Trajectory Classification , 2012, 2012 IEEE 28th International Conference on Data Engineering.

[17]  Yu Zheng,et al.  Travel time estimation of a path using sparse trajectories , 2014, KDD.

[18]  Mehdi Boukhechba,et al.  Online recognition of people's activities from raw GPS data: semantic trajectory data analysis , 2015, PETRA.

[19]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[20]  Sotiris B. Kotsiantis,et al.  Supervised Machine Learning: A Review of Classification Techniques , 2007, Informatica.

[21]  Zheping Shao,et al.  Maritime Anomaly Detection within Coastal Waters Based on Vessel Trajectory Clustering and Naïve Bayes Classifier , 2017, Journal of Navigation.

[22]  Peter Harrington,et al.  Machine Learning in Action , 2012 .

[23]  Dhaval Patel.  Incorporating duration and region association information in trajectory classification , 2013, J. Locat. Based Serv..