Skip to main navigation Skip to search Skip to main content

Active tracking strategy with multiple cameras in large areas

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

Abstract

This paper presents a distributed multiple camera collaboration strategy for active tracking in large areas. Their collaboration utilizes a sector-based mechanism in which the visual correspondences among different cameras are determined. In order to emulate the master-slave operations, the proposed method combines the local data to construct the global information. Based on the global information, the loads are evenly distributed to avoid tracking misses due to the dynamic behaviors of the objects. The trajectories of all objects visible in the local cameras are estimated for actual tracking and the estimated dynamics are used for scheduling of the cameras. The active tracking triggering timing is carefully chosen to maximize the overall monitoring time for general surveillance operations.

Original languageEnglish
Title of host publicationAdvances in Computer Science and Ubiquitous Computing - CSA-CUTE2016
EditorsVincenzo Loia, James J. Jong Hyuk Park, Gangman Yi, Yi Pan
PublisherSpringer Verlag
Pages686-693
Number of pages8
ISBN (Print)9789811030222
DOIs
StatePublished - 2017
Event11th International Conference on Ubiquitous Information Technologies and Applications, CUTE 2016 - Bangkok, Thailand
Duration: Dec 19 2016Dec 21 2016

Publication series

NameLecture Notes in Electrical Engineering
Volume421

Conference

Conference11th International Conference on Ubiquitous Information Technologies and Applications, CUTE 2016
Country/TerritoryThailand
CityBangkok
Period12/19/1612/21/16

Keywords

  • Active tracking
  • Camera collaboration
  • Surveillance systems

Fingerprint

Dive into the research topics of 'Active tracking strategy with multiple cameras in large areas'. Together they form a unique fingerprint.

Cite this