A decision level fusion of morphology based Pattern Spectrum and IDSC for shape representation and classification

Abstract

In this paper, we propose a combined classifier approach based on Pattern Spectrum (PS) and Inner Distance Shape Context (IDSC) to classify shapes accurately. The PS captures the intrinsic details of shape and the inner-distance is insensitive to shape articulations, and hence combined to represent/classify shapes accurately. The Earth Movers Distance (EMD) metric in case of PS and Dynamic Programming (DP) in case of IDSC were respectively employed to obtain similarity values and fused to classify a given query shape. The experiments are conducted on publicly available shape datasets namely MPEG-7, Kimia-99, Kimia-216, Myth and Tools-2D and the results are presented by means of Bull's eye score. The comparative study is also provided with the well known approaches to exhibit the retrieval accuracy of the proposed approach. The experimental results demonstrate that the proposed approach yields significant improvements over baseline shape matching algorithms.