Shape recognition from large image libraries by inexact graph matching

This paper describes a graph-matching technique for recognising line-patter n shapes in large image databases. The methodological contribution of the p aper is to develop a Bayesian matching algorithm that uses edge-consistency and node attribute similarity. This information is used to determine the a posteriori probability of a query graph for each of the candidate matches in the database. The node feature-vectors are constructed by computing norm alised histograms of pairwise geometric attributes. Attribute similarity is assessed by computing the Bhattacharyya distance between the histograms. R ecognition is realised by selecting the candidate from the database which h as the largest a posteriori probability. (C) 1999 Elsevier Science B.V. All rights reserved.