• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to project page

2023 Fiscal Year Final Research Report

Next-generation intelligent security camera systems that predict criminal event

Research Project

  • PDF
Project/Area Number 20H02400
Research Category

Grant-in-Aid for Scientific Research (B)

Allocation TypeSingle-year Grants
Section一般
Review Section Basic Section 25020:Safety engineering-related
Research InstitutionUniversity of the Ryukyus

Principal Investigator

Nagayama Itaru  琉球大学, 工学部, 教授 (80274885)

Project Period (FY) 2020-04-01 – 2024-03-31
Keywords画像処理 / 防犯カメラ / 機械学習 / ひったくり / 人工知能 / 知能情報処理 / モーション予測 / 挙動解析
Outline of Final Research Achievements

In this study、 research and development of a next-generation intelligent security camera that predicts and detects the occurrence of street crimes such as purse-snatching was carried out. The occurrence of a crime is predicted by AI and machine learning technology based on recent behaviour in security camera images、 and a loud alarm is issued just before the occurrence of a crime. This is expected to prevent the offender from committing the crime.
The results of the study shows that the accuracy of predicting the occurrence of purse-snatching 500 milliseconds in advance was 94.25%、 followed by an accuracy of 86.75% when predicting the occurrence 600 milliseconds in advance.

Free Research Field

知能情報工学

Academic Significance and Societal Importance of the Research Achievements

これまで我々はリアルタイムに発生した「ひったくり」「拉致誘拐」等の犯罪行為をAI・機械学習技術を用いて自動検知する次世代型知的防犯カメラシステムの研究を行ってきた。ところが、これらの研究はひったくり等の状況を自動検知し、警察・救急への自動通報を実行できるが、犯行や事件発生そのものを防ぐことはできないという根本的限界があった。そのため、「ひったくり」の犯行発生を数秒前に予知・予測し、犯行直前に大音量警報アラームを発報する超次世代知的防犯カメラシステムの研究に取組んだ。将来的に、犯行の発生直前に予防的介入を行うことで犯行そのものを防止する先進的システムの開発が期待される。

URL: 

Published: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi