Morphological Segmentation of Binary Patterns

Abstract

This paper presents a method for segmenting binary patterns into seven mutually exclusive categories: core, islet, loop, bridge, perforation, edge, and branch. This is achieved by applying a series of morphological transformations such as erosions, geodesic dilations, reconstruction by dilation, anchored skeletonisation, etc. The proposed method depends on a single parameter only and can be used for characterising binary patterns with emphasis on connections between their parts as measured at varying analysis scales. This is illustrated on two examples related to land cover maps and circuit board defect detection.