Gözetleme videolarinda artimli negatif olmayan matris ayriştirma ile arka plan modelleme

Translated title of the contribution: Incremental nonnegative matrix factorization for background modeling in surveillance video

Serhat S. Bucak, Bilge Günsel, Ozan Gürsoy

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

9 Citations (Scopus)

Abstract

In this paper, we propose an Incremental Non-negative Matrix Factorization (INMF) method which can be effectively used for dynamic background modeling in surveillance applications. The proposed factorization method is derived from Non-negative Matrix Factorization (NMF), and models the dynamic content of the video by controlling contribution of the subsequent observations to the existing model adaptively. Unlike the batch nature of NMF, INMF is an on-line content representation scheme which is capable of extracting moving foreground objects. Test results are reported in order to compare background modeling performances of INMF, NMF and Incremental Principal Components Analysis (IPCA). It is concluded that INMF outperforms both NMF and IPCA and its robustness to illumination changes makes it as a powerful representation tool in surveillance applications.

Translated title of the contributionIncremental nonnegative matrix factorization for background modeling in surveillance video
Original languageTurkish
Title of host publication2007 IEEE 15th Signal Processing and Communications Applications, SIU
DOIs
Publication statusPublished - 2007
Event2007 IEEE 15th Signal Processing and Communications Applications, SIU - Eskisehir, Turkey
Duration: 11 Jun 200713 Jun 2007

Publication series

Name2007 IEEE 15th Signal Processing and Communications Applications, SIU

Conference

Conference2007 IEEE 15th Signal Processing and Communications Applications, SIU
Country/TerritoryTurkey
CityEskisehir
Period11/06/0713/06/07

Fingerprint

Dive into the research topics of 'Incremental nonnegative matrix factorization for background modeling in surveillance video'. Together they form a unique fingerprint.

Cite this