Shock Graph for Shape Recognition

Shock graphs have emerged as a powerful generic 2-D shape representation. In this paper, we propose a framework to address the problem of generic 2-D shape recognition. The aim is to use shock graphs as shape representation for recognition where features of objects are detected for recognition and classification. We propose to represent the medial axis characteristic points as an attribute to model the shape. The information about the object shape and its topology is totally embedded in them and this allows the potential to compare different shapes by their shock graph. The techniques developed are skeletonization, followed by thinning and finally detection of end points and junction points that are present in every region of the shock graph for possible modelling of shape recognition. Experimental results demonstrate the correctness in detecting its characteristic points for recognition. It will be shown that this framework is effective in shape recognition. The algorithms were tested with wide varieties of 2D shapes. Research on this promising approach is ongoing, focusing both on new applications and on techniques that will further enhance its performance in a variety of computer vision and pattern recognition domains.