A Method of Detecting Abnormal Crowd Behavior Events Applied in Air Patrol Robot

DOI
  • Zhao Huailin
    School of Electrical and Electronic Engineering ,Shanghai Institute of Technology
  • Wang Shunzhou
    School of Electrical and Electronic Engineering ,Shanghai Institute of Technology
  • Xu Shifang
    School of Electrical and Electronic Engineering ,Shanghai Institute of Technology
  • Zhang Yani
    School of Computer Science and Information Engineering ,Shanghai Institute of Technology
  • Sugisaka Masanori
    ALife Robotics Corporation Ltd.

説明

When the ground or air patrol robot monitors a certain area, one of the important intelligent functions is to estimate the crowd density of the monitored area. This paper analyzes the crowd density estimation algorithm, and use a Gaussian process regression model for crowd density estimation. Through the crowd density estimation and changes, we can detect abnormal behavior events of the crowd. The method can not only estimate the population density of the specified area, but also analyze and detect the abnormal behavior events of the crowd. This application provides an important technical support for enhancing the patrol robot monitoring effect.

収録刊行物

キーワード

詳細情報 詳細情報について

  • CRID
    1390845713072099200
  • DOI
    10.5954/icarob.2017.os16-1
  • ISSN
    21887829
  • 本文言語コード
    en
  • データソース種別
    • JaLC
    • Crossref
  • 抄録ライセンスフラグ
    使用不可

問題の指摘

ページトップへ