Fuzzy Models for Low-Level Computer Vision: A Comprehensive Approach

It is well-known that fuzzy sets were conceived by Zadeh in 1965 as a mathematical tool able to model the concept of partial membership. After a period of theoretical investigation, in the mid- 1980s fuzzy rule-based methods became a problem solving technology and the engineering applications grew fast especially in the area of control systems. Low-level computer vision was a field where fuzzy modelling emerged as a very powerful resource too. The aim of this presentation is not to provide a thorough description of many different approaches that are currently available in the scientific literature. It aims rather at investigating how nowadays key operations such as noise removal, image sharpening and edge detection can be performed by adopting a comprehensive approach and simple fuzzy models.