Hierarchical Tree Structure Based 2D Shape Matching

Abstract

This paper presents 2D shape matching technique using Hierarchical tree structure extracted from shock graph that in turn extracted from the skeleton of the shape of interest. Object recognition and shape matching are important issues in the field of image processing. Extraction and application of skeleton of a shape is widely used in these fields. In this paper shape matching and retrieval is discussed, using one of the skeleton-based methods called “shock graphs”. Missing parts and presence of occlusion in the shapes can not guarantee better results with older techniques. But by modifying and optimizing this “shock graph” method, results may improve significantly. The process of 2D shape matching using hierarchical tree structures contain mainly three steps such as the skeleton formation from the 2D silhouette, converting the skeleton into its corresponding shock graphs and the tree like hierarchical data structure and finally comparing so obtained tree with the other tree structures in the database for recognition. This particular approach for shape matching is memory efficient as the lesser details of the shape of interest need to be stored in the database for future matching.