Fuzzy Decision Maker for Knowledge Discovery from Image Archives

Abstract

With this enormous speed in generating and collecting images, there is an extreme need in extracting interesting and useful knowledge from image archives. In our previous works, we have proposed an image mining framework to extract knowledge from a sequence of images. The framework is composed of two main modules: image analysis and knowledge processing. In this paper, we successfully customized the knowledge processing module to check legality/normality of vehicles/people behaviors in different image sequences by means of a fuzzy decision maker (FDM).