March 2008

Conference Paper

Multiple target detection in video using quadratic multi-frame correlation filtring

By:
Kerekes, Ryan A; Kumar, B. V. K. Vijaya
Volume:
6977
Publication Date:
March 2008
Conference Name:
22nd Congress of the International Commission for Optics: Light for the Development of the World
Conference Location:
Orlando, Florida, United States of America

Abstract

Most integrated target detection and tracking systems employ state space models to keep track of an explicit number of individual targets.Recently,anon-state-spaceframeworkwasdevelopedforenhancingtargetde-tectioninvideobyapplyingprobabilisticmotionmodelstothesoftinformationincorrelationoutputsbeforethresholding.Thisframeworkhasbeenreferredtoasmulti-framecorrelationfiltering(MFCF),andbecauseitavoidstheuseofstate-spacemodelsandtheformationofexplicittracks,theframeworkiswell-suitedforhan-dlingsceneswithunknownnumbersoftargetsatunknownpositions.Inthispaper,weproposetousequadraticcorrelationfilters(QCFs)intheMFCFframeworkforrobusttargetdetection.Wetestourdetectionalgorithmonrealandsynthesizedsingle-targetandmulti-targetvideosequences.SimulationresultsshowthatMFCFcansignificantlyreduce(tozerointhebestcase)thefalsealarmratesofQCFsatdetectionratesabove95%inthepresenceoflargeamountsofuncorrelatednoise.WealsoshowthatMFCFismoreadeptatrejectingthosefalsepeaksduetouncorrelatednoiseratherthanthoseduetoclutterandcompressionnoise;consequently,weshowthatfiltersusedintheframeworkshouldbemadetofavorclutterrejectionovernoisetolerance.


Related Researchers