A unified model for context-based behavioural modelling and classification

作者：

Highlights：

• A DBN for behavioural modelling, information fusion and classification is proposed.

• Classification is performed using the Gaussian sum filter inference algorithm.

• The proposed model is applied to the problem of maritime piracy.

• Pirate, transport and fishing vessels are classified according to their behaviour.

摘要

•A DBN for behavioural modelling, information fusion and classification is proposed.•Classification is performed using the Gaussian sum filter inference algorithm.•The proposed model is applied to the problem of maritime piracy.•Pirate, transport and fishing vessels are classified according to their behaviour.

论文关键词：Dynamic Bayesian network,Switching linear dynamical system,Information fusion,Behaviour modelling,Activity recognition,Maritime piracy

论文评审过程：Available online 5 May 2015, Version of Record 26 May 2015.

论文官网地址：https://doi.org/10.1016/j.eswa.2015.04.061