Learning and Recognizing Behavioral Patterns based on Comprehensive Movement of People

2015 Fiscal Year Final Research Report

• Project/Area Number: 24700195
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Perception information processing/Intelligent robotics
• Research Institution: Osaka Prefecture University
• Principal Investigator: AOKI Shigeki 大阪府立大学, 現代システム科学域, 准教授 (90390541)
• Project Period (FY): 2012-04-01 – 2016-03-31
• Keywords: 大まかな動き / 行動パターン / 非日常状態検出 / 監視カメラ

Outline of Final Research Achievements

Recently, automatic video surveillance systems have become an active area of research. There are detecting irregular behavior based on a set of given irregular behavior models, and detecting rare behavior as irregular in existing methods. However, since these methods use specific features to each environment, it is not possible to apply learning result in the environment to other similar environment without relearning behavioral patterns.
We propose a novel method to detect irregular situations in similar multiple environments. So our method uses common features in some environment, the existing problem of relearning behavioral patterns is resolved.

Free Research Field

知覚情報処理・知能ロボティクス