Shape understanding system: the application of fuzzy sets, neural networks and statistical methods in the process of visual thinking

摘要

The application of fuzzy sets, neural networks and statistical methods in a shape understanding system (SUS), is presented. The shape understanding system operates based on the knowledge of image processing, decision making and search strategies as well as the knowledge of shape description and representation distributed among the specialized experts. The SUS is built to imitate the capability of the visual system in performing visual thinking.