Behaviour Analysis and Prediction in Image Sequences using Rough Sets

摘要

In this paper, a rough set based data mining technique is extended and applied to learn behaviour patterns of moving objects in videos. An intelligent image analysis system is proposed to discover knowledge from video collections. The system is applied to synthetic image sequences containing motions of a predator and prey behaving according to a set of rules. The results show that the system is able to discover the rules governing the predator behaviour and is also able to predict the predator's final decision using the rules discovered.