Fuzzy classification of generic edge features

This paper presents a fuzzy logic approach for classifying generic edge segments (GES). A set of GES is modeled in descriptive geometry based on perceptual organization principles. Each GES represents a class of edge segments which belong to a same perceptual group and measured according to the generic criteria of perceptual organization laws. A fuzzy logic system of the CES models was developed which improves the robustness and accuracy of GES classification. In the proposed approach both the GES representation and extraction are handled qualitatively. The experiment results demonstrate the effectiveness of the proposed technology for edge feature classification and extraction which can be beneficial for many image analysis applications.