Omni-directional vision sensor based behavior monitoring system using Bayesian Network

Abstract

This paper describes a behavior monitoring system based on an omni-directional vision sensor as one of the important elements in realizing "Sensing and Robotic Support Room" for elderly people. Such support rooms are expected to be further developed in the future with the high performance to automatically understand the human's actions and behavior patterns and detect the unusual patterns using some sensors and to support their daily lives and motions using some robotic control systems. The proposed monitoring system based on an omni-directional vision sensor automatically learns the daily behavior patterns and detects the unusual behavior patterns and actions using Bayesian Network representation. The Bayesian Network is constructed using image feature values extracted from the captured image sequence and the respective behavior patterns are represented as the conditional probabilities. Unusual behavior patterns can be automatically detected based on the generation probabilities using Neural Network analysis. Some experiments show the effectiveness of the proposed system.