A fast algorithm for adaptive background model construction using Parzen density estimation

Tatsuya Tanaka, Daisaku Arita, Atsushi Shimada, Rin Ichiro Taniguchi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

37 Citations (Scopus)

Abstract

Non-parametric representation of pixel intensity distribution is quite effective to construct proper background model and to detect foreground objects accurately. However, from the viewpoint of practical application, the computation cost of the distribution estimation should be reduced. In this paper, we present fast estimation of the probability density function (PDF) of pixel value using Parzen density estimation and foreground object detection based on the estimated PDF. Here, the PDF is computed by partially updating the PDF estimated at the previous frame, and it greatly reduces the computation cost of the PDF estimation. Thus, the background model adapts quickly to changes in the scene and, therefore, foreground objects can be robustly detected. Several experiments show the effectiveness of our approach.

Original languageEnglish
Title of host publication2007 IEEE Conference on Advanced Video and Signal Based Surveillance, AVSS 2007 Proceedings
Pages528-533
Number of pages6
DOIs
Publication statusPublished - 2007
Event2007 IEEE Conference on Advanced Video and Signal Based Surveillance, AVSS 2007 - London, United Kingdom
Duration: Sept 5 2007Sept 7 2007

Publication series

Name2007 IEEE Conference on Advanced Video and Signal Based Surveillance, AVSS 2007 Proceedings

Other

Other2007 IEEE Conference on Advanced Video and Signal Based Surveillance, AVSS 2007
Country/TerritoryUnited Kingdom
CityLondon
Period9/5/079/7/07

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Signal Processing

Fingerprint

Dive into the research topics of 'A fast algorithm for adaptive background model construction using Parzen density estimation'. Together they form a unique fingerprint.

Cite this