Image Micro-pattern Analysis Using Fuzzy Numbers

This paper proposes a new methodology for micro pattern analysis in digital images based on fuzzy numbers. A micro-pattern is the structure of the gray-level pixels within a neighborhood and can describe the spatial context of the image, such as edge, line, spot, blob, corner, texture, and more complex patterns. By treating a pixel neighborhood as a fuzzy set and each pixel gray-level as a fuzzy number, we can evaluate the membership degree of the central pixel to the others. We have called this method the Local Fuzzy Pattern (LFP). Using a sigmoid membership function, we proved that the proposed methodology surpasses the Hit-rate of the Local Binary Pattern (LBP), for texture classification. The LFP proved to be robust to image rotation. Moreover, our proposed formulation for the LFP is a generalization of previously published techniques, such as Texture Unit, LBP, FUNED, and Census Transform.