Dissecting uncertainty-based fusion techniques for maritime anomaly detection

Detecting and classifying anomalies for Maritime Situation Awareness gets a lot of benefit from the combination of multiple sources, correlating their output for detecting inconsistencies in vessels' behaviour. Adequate uncertainty representation and processing is crucial for this higher-level task where the operator analyses information correlating with his background knowledge. This paper addresses the problem of performance criteria selection and definition for information fusion systems in their ability to handle uncertainty. Indeed, i addition to the classical algorithmic performances of accuracy or timeliness, other aspects such as the interpretation, simplicity, expressiveness need to be considered in the design of the technique for uncertainty management for a improved synergy between the human and the system. In this paper, we dissect several uncertainty representation and reasoning techniques (URRTs) addressing a fusion problem for maritime anomaly detection. The uncertainty supports are identified as a basis for the global expressiveness criterion. A selection of six elementary URRTs are described and compared according to their expressiveness power of uncertainty, using the Uncertainty Representation and Reasoning Framework (URREF) ontology. This study is considered as preliminary to guide further development and implementation of fusion algorithms for maritime anomaly detection, and the definition of associated criteria and measures of performance.

[1]  Kalyan Moy Gupta,et al.  A comparative evaluation of anomaly detection algorithms for maritime video surveillance , 2011, Defense + Commercial Sensing.

[2]  Johan Pieter de Villiers,et al.  A URREF interpretation of Bayesian network information fusion , 2014, 17th International Conference on Information Fusion (FUSION).

[3]  Richard O. Lane,et al.  Maritime anomaly detection and threat assessment , 2010, 2010 13th International Conference on Information Fusion.

[4]  Rikard Laxhammar,et al.  Anomaly detection for sea surveillance , 2008, 2008 11th International Conference on Information Fusion.

[5]  Thierry Denoeux,et al.  Refined modeling of sensor reliability in the belief function framework using contextual discounting , 2008, Inf. Fusion.

[6]  Kalyan Moy Gupta,et al.  Case-Based Collective Inference for Maritime Object Classification , 2009, ICCBR.

[7]  Philippe Smets,et al.  Classification Using Belief Functions: Relationship Between Case-Based and Model-Based Approaches , 2006, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[8]  Ronnie Johansson,et al.  Characterization and Empirical Evaluation of Bayesian and Credal Combination Operators , 2011, J. Adv. Inf. Fusion.

[9]  Erik Blasch,et al.  Towards unbiased evaluation of uncertainty reasoning: The URREF ontology , 2012, 2012 15th International Conference on Information Fusion.

[10]  Audun Jøsang,et al.  URREF self-confidence in information fusion trust , 2014, 17th International Conference on Information Fusion (FUSION).

[11]  Rikard Laxhammar,et al.  Anomaly Detection in Trajectory Data for Surveillance Applications , 2011 .

[12]  Branko Ristic,et al.  Classification with imprecise likelihoods: A comparison of TBM, random set and imprecise probability approach , 2011, 14th International Conference on Information Fusion.

[13]  P. Smets,et al.  Target classification approach based on the belief function theory , 2005, IEEE Transactions on Aerospace and Electronic Systems.

[14]  Philippe Smets,et al.  Belief functions: The disjunctive rule of combination and the generalized Bayesian theorem , 1993, Int. J. Approx. Reason..

[15]  Josef Kittler,et al.  Combining classifiers: A theoretical framework , 1998, Pattern Analysis and Applications.

[16]  P. Smets.  Imperfect information : Imprecision-Uncertainty , 1999 .

[17]  Henry Leung,et al.  Bayesian and Dempster-Shafer target identification for radar surveillance , 2000, IEEE Trans. Aerosp. Electron. Syst..

[18]  Jerome H. Friedman,et al.  On Bias, Variance, 0/1—Loss, and the Curse-of-Dimensionality , 2004, Data Mining and Knowledge Discovery.

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

[20]  Philippe Smets,et al.  Imperfect Information: Imprecision and Uncertainty , 1996, Uncertainty Management in Information Systems.

[21]  Weiqiang Dong.  On Bias , Variance , 0 / 1-Loss , and the Curse of Dimensionality RK April 13 , 2014 .

[22]  F. Johansson,et al.  Detection of vessel anomalies - a Bayesian network approach , 2007, 2007 3rd International Conference on Intelligent Sensors, Sensor Networks and Information.

[23]  Michele Vespe,et al.  Vessel Pattern Knowledge Discovery from AIS Data: A Framework for Anomaly Detection and Route Prediction , 2013, Entropy.