Real-time people counting method with surveillance cameras implemented on embedded system

Hsiang Chieh Chen, Ya Ching Chang, Nai Jen Li, Cheng Feng Weng, Wen June Wang

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

2 Scopus citations

Abstract

This work presents an image-based method to estimate the number of people crowd in a large-spaced indoor environment. The method is implemented on embedded system using a digital signal processor. Moving people in each video frame are first extracted using background subtraction method. Additionally, an adaptation scheme is employed to update background model for conditions of sudden lighting changes and abandoned objects. The extracted foreground objects are further accumulated as weighted pixels to estimate the number of people, where the weights are automatically calculated by considering camera settings. Experimental results demonstrate that the proposed method performs well on estimating the number of people in the video surveillance.

Original languageEnglish
Title of host publicationWCECS 2013 - World Congress on Engineering and Computer Science 2013
PublisherNewswood Limited
Pages512-515
Number of pages4
ISBN (Print)9789881925169
StatePublished - 2013
Event2013 World Congress on Engineering and Computer Science, WCECS 2013 - San Francisco, CA, United States
Duration: 23 Oct 201325 Oct 2013

Publication series

NameLecture Notes in Engineering and Computer Science
Volume1
ISSN (Print)2078-0958

Conference

Conference2013 World Congress on Engineering and Computer Science, WCECS 2013
Country/TerritoryUnited States
CitySan Francisco, CA
Period23/10/1325/10/13

Keywords

  • Digital signal processor
  • People counting
  • Video surveillance

Fingerprint

Dive into the research topics of 'Real-time people counting method with surveillance cameras implemented on embedded system'. Together they form a unique fingerprint.

Cite this