%0 Conference Proceedings %T A Product graph based method for dual subgraph matching applied to symbol spotting %A Anjan Dutta %A Josep Llados %A Horst Bunke %A Umapada Pal %B 10th IAPR International Workshop on Graphics Recognition %D 2013 %F Anjan Dutta2013 %O DAG %O exported from refbase (http://refbase.cvc.uab.es/show.php?record=2359), last updated on Thu, 10 Nov 2016 12:06:21 +0100 %X Product graph has been shown to be an efficient way for matching subgraphs. This paper reports the extension of the product graph methodology for subgraph matching applied to symbol spotting in graphical documents. This paper focuses on the two major limitations of the previous version of product graph: (1) Spurious nodes and edges in the graph representation and (2) Inefficient node and edge attributes. To deal with noisy information of vectorized graphical documents, we consider a dual graph representation on the original graph representing the graphical information and the product graph is computed between the dual graphs of the query graphs and the input graph.The dual graph with redundant edges is helpful for efficient and tolerating encoding of the structural information of the graphical documents. The adjacency matrix of the product graph locates similar path information of two graphs and exponentiating the adjacency matrix finds similar paths of greater lengths. Nodes joining similar paths between two graphs are found by combining different exponentials of adjacency matrices. An experimental investigation reveals that the recall obtained by this approach is quite encouraging. %U http://refbase.cvc.uab.es/files/DLB2013b.pdf