Skip to main navigation Skip to search Skip to main content

Surveillance system of power transmission line via object recognition and 3D vision computation

  • Xi'an Jiaotong University

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

Abstract

Surveillance systems have been widely applied on power transmission system for security precaution of the major risk factor that is the construction activity in the vicinity of power transmission line. However, currently used automatic object detection in surveillance systems suffers from high error rate and has at least two limitations: first, the type of the object can™t be recognized; second, the dangerous strength of the object cannot be identified. In this paper, we propose a video surveillance method for the security precaution of power transmission line via the techniques of object recognition and 3D spatial location detection so that the motion objects are recognized and the position and size of the object are determined to identify the dangerous strength. Experimental results show that the developed system based on our proposed method is feasible and practical.

Original languageEnglish
Title of host publicationProceedings of SPIE-IS and T Electronic Imaging - Digital Photography X
PublisherSPIE
ISBN (Print)9780819499400
DOIs
StatePublished - 2014
EventDigital Photography X - San Francisco, CA, United States
Duration: 3 Feb 20145 Feb 2014

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume9023
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

ConferenceDigital Photography X
Country/TerritoryUnited States
CitySan Francisco, CA
Period3/02/145/02/14

Keywords

  • 3D vision computation
  • object recognition
  • power transmission line
  • security precaution
  • surveillance system

Fingerprint

Dive into the research topics of 'Surveillance system of power transmission line via object recognition and 3D vision computation'. Together they form a unique fingerprint.

Cite this