An effective method for detecting snatch thieves in video surveillance

DOI オープンアクセス

説明

Nowadays, a tremendous amount of accidents and terrorisms has been occurred all over the world no exception in Japan. Thus, detection of suspicious activities in public areas such as railway stations, shopping malls and many other areas using video surveillance becomes important. However, very little has been achieved regarding real-time event recognition of two person interactions such as snatch theft events. Moreover the way of snatch theft has been increasing like using a motorcycle and so on. In this paper, we propose an effective method for detecting the snatch theft event between two persons. Specifically, the proposed method consists of several steps: pedestrian tracking, feature computation and snatch theft detection. For feature computation, shape and motion features are used. To confirm the validity of proposed method some experimental results in a various situations are shown by using some collected video sequences.

収録刊行物

キーワード

詳細情報 詳細情報について

  • CRID
    1390564238096155776
  • DOI
    10.5954/icarob.2017.os20-1
  • ISSN
    21887829
  • 本文言語コード
    en
  • データソース種別
    • JaLC
    • Crossref
    • OpenAIRE
  • 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