DocumentCode :
549007
Title :
Hybrid intensity and likelihood ratio tracking (iLRT) filter for multitarget detection
Author :
Streit, Roy ; Osborn, Bryan ; Orlov, Kirill
Author_Institution :
Metron, Reston, VA, USA
fYear :
2011
fDate :
5-8 July 2011
Firstpage :
1
Lastpage :
8
Abstract :
A particle version of a PHD / Intensity filter is coupled with a cumulative log-likelihood ratio (CLLR) detector to enable multiple target detection and track extraction. The filter provides each particle with a path history, or target trajectory. The CLLR evaluated for such a particle path is the quickest possible detector of target presence on the path. This hybrid approach is useful because track extraction methods based on the CLLR for the particle paths are grounded in an optimal target detection statistic, not merely on the relative abundance of particles that represent the output of the PHD / Intensity filter. The utility of the hybrid intensity and likelihood ratio tracking (iLRT) filter approach is demonstrated by example.
Keywords :
maximum likelihood detection; object detection; particle filtering (numerical methods); tracking filters; CLLR detector; PHD/intensity filter; cumulative log-likelihood ratio; hybrid intensity; iLRT filter; likelihood ratio tracking filter; multiple target detection; multitarget detection; particle version; target trajectory; track extraction; Clutter; Feature extraction; Imaging; Joints; Kinematics; Object detection; Target tracking; Cumulative Loglikelihood Ratio; Intensity Filter; Likelihood Ratio Tracking; MSTWG; Multistatic Sonar; Multitarget Tracking; PHD Filter; Particle Filter; Track Heritage; iLRT;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Information Fusion (FUSION), 2011 Proceedings of the 14th International Conference on
Conference_Location :
Chicago, IL
Print_ISBN :
978-1-4577-0267-9
Type :
conf
Filename :
5977441
Link To Document :
بازگشت