An Improved Shock Graph for an Improved Object Recognition

Converting a binary image to a skeleton or medial axis form is often used to preserve the shape details efficiently. The medial axis is converted to a shock graph which has structure like a tree. Shock graphs are derived from the skeleton and have emerged as powerful 2-D shape representation. A skeleton has number of branches. A branch is a connected set of points between an end point and a joint or another end point. Every point also called as shock point on a skeleton is labeled according to the variation of the radius function. The labeled points in a given branch are to be grouped according to their labels and connectivity, so that each group of same-label connected points will be stored in a graph node. Edges are added between the nodes so as to produce a directed acyclic graph. Binary images with different shapes have different skeleton and different tree structure. Number of features of the graph can be extracted which facilitate comparison of shapes using these features. Comparison of shapes using their Shock graphs provides a very effective way of object recognition. An object recognition frame work by comparing the subgraphs has been presented here. A novel concept has been implemented in the presented work for inserting a node in Shock branch whenever there is a sharp direction change. Addition of node improves the object recognition results. Keywords— Directed acyclic graph, MAT, Radius function,Skeleton,Shock graph, nodes, labels, shock graph grammar.