Unsupervised track classification based on Hierarchical Dirichlet Processes

Author

Sildam, Juri ; Braca, Paolo ; Lepage, Kyle ; Willett, P.

Author_Institution

NATO STO CMRE, La Spezia, Italy

fYear

2013

fDate

9-13 Sept. 2013

Firstpage

1

Lastpage

5

Abstract

An unsupervised track classification approach is applied to sonar multistatic multitarget tracking. Appropriate discriminative and aggregative features are derived from beamformed and normalized matched-filtered data as recorded from a linear array towed behind an AUV. A clustering algorithm based on Hierarchical Dirichlet Processes is proposed for unsupervised classification of tracks. Overall improvement of target tracking is demonstrated via the Optimal Subpattern Assignment metric.

Keywords

array signal processing; electrical engineering computing; matched filters; pattern clustering; signal classification; sonar signal processing; sonar tracking; statistical distributions; target tracking; unsupervised learning; AUV; aggregative features; beamforming; clustering algorithm; discriminative features; hierarchical Dirichlet processes; linear array; normalized matched-filtered data; optimal subpattern assignment metric; sonar multistatic multitarget tracking; unsupervised track classification approach; Entropy; Hidden Markov models; Sea measurements; Signal to noise ratio; Target tracking; Trajectory; clustering; hierarchical Dirichlet processes; infinite HMM; tracking;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing Conference (EUSIPCO), 2013 Proceedings of the 21st European

Conference_Location

Marrakech

Type

conf

Filename

6811817