DocumentCode :
567571
Title :
Maritime anomaly detection using Gaussian Process active learning
Author :
Kowalska, Kira ; Peel, Leto
Author_Institution :
Adv. Technol. Centre, BAE Syst., Bristol, UK
fYear :
2012
fDate :
9-12 July 2012
Firstpage :
1164
Lastpage :
1171
Abstract :
A model of normal vessel behaviours is useful for detecting illegal, suspicious, or unsafe behaviour; such as vessel theft, drugs smuggling, people trafficking or poor sailing. This work presents a data-driven non-parametric Bayesian model, based on Gaussian Processes, to model normal shipping behaviour. This model is learned from Automatic Identification System (AIS) data and uses an Active Learning paradigm to select an informative subsample of the data to reduce the computational complexity of training. The resultant model allows a measure of normality to be calculated for each newly-observed transmission according to its velocity given its current latitude and longitude. Using this measure of normality, ships can be identified as potentially anomalous and prioritised for further investigation. The model performance is assessed by its ability to detect artificially generated AIS anomalies at locations around the United Kingdom. Finally, the model is demonstrated on case studies from artificial and real vessel data to detect anomalies in unusual tracks.
Keywords :
Bayes methods; Gaussian processes; behavioural sciences; computational complexity; data mining; identification technology; learning (artificial intelligence); sampling methods; training; AIS data; Gaussian process active learning; United Kingdom; artificial vessel data; artificially generated AIS anomalies detection; automatic identification system data; data-driven nonparametric Bayesian model; illegal behaviour detection; informative data subsampling; maritime anomaly detection; newly-observed transmission; normal shipping behaviour model; normal vessel behaviours; real vessel data; suspicious behaviour detection; training computational complexity; unsafe behaviour detection; Accuracy; Bayesian methods; Computational modeling; Data models; Gaussian processes; Marine vehicles; Training;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Information Fusion (FUSION), 2012 15th International Conference on
Conference_Location :
Singapore
Print_ISBN :
978-1-4673-0417-7
Electronic_ISBN :
978-0-9824438-4-2
Type :
conf
Filename :
6289940
Link To Document :
بازگشت