HMM-based graph edit distance for image indexing

Most of the existing graph edit distance (GED) algorithms require cost functions which are difficult to be defined exactly. In this article, we propose a cost function free algorithm for computing GED. It only depends on the distribution of nodes rather than node or edge attributes in graphs. Hidden Markov model (HMM) is employed to model the distribution of feature points and thus dissimilarity measure of graphs can be posed as distance of HMMs. A fast algorithm of Kullback-Leibler Distance, suitable for computing the distance between two probability models, is adopted to compute the distance of HMMs. Experimental results demonstrate that the proposed GED algorithm can characterize the structure variety of graphs effectively and is available for clustering and indexing images of both rigid and nonrigid bodies. © 2008 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 18, 209–218, 2008

[1]  Stephen M. Smith,et al.  SUSAN—A New Approach to Low Level Image Processing , 1997, International Journal of Computer Vision.

[2]  Edwin R. Hancock,et al.  Structural Matching by Discrete Relaxation , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  S. C. Johnson Hierarchical clustering schemes , 1967, Psychometrika.

[4]  H Qian Relative entropy: free energy associated with equilibrium fluctuations and nonequilibrium deviations. , 2001, Physical review. E, Statistical, nonlinear, and soft matter physics.

[5]  Horst Bunke,et al.  On a relation between graph edit distance and maximum common subgraph , 1997, Pattern Recognit. Lett..

[6]  Xuelong Li,et al.  General Tensor Discriminant Analysis and Gabor Features for Gait Recognition , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[7]  Horst Bunke,et al.  Recent developments in graph matching , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[8]  Hans P. Moravec Obstacle avoidance and navigation in the real world by a seeing robot rover , 1980 .

[9]  M. Do Fast approximation of Kullback-Leibler distance for dependence trees and hidden Markov models , 2003, IEEE Signal Processing Letters.

[10]  Horst Bunke,et al.  A New Algorithm for Error-Tolerant Subgraph Isomorphism Detection , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[11]  Kaizhong Zhang,et al.  A constrained edit distance between unordered labeled trees , 1996, Algorithmica.

[12]  Edwin R. Hancock,et al.  Bayesian Graph Edit Distance , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[13]  King-Sun Fu,et al.  A distance measure between attributed relational graphs for pattern recognition , 1983, IEEE Transactions on Systems, Man, and Cybernetics.

[14]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[15]  Antonio Robles-Kelly,et al.  Graph edit distance from spectral seriation , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  Enrique Vidal,et al.  Computation of Normalized Edit Distance and Applications , 1993, IEEE Trans. Pattern Anal. Mach. Intell..

[17]  Vladimir I. Levenshtein,et al.  Binary codes capable of correcting deletions, insertions, and reversals , 1965 .

[18]  Ferdinando Silvestro Samaria,et al.  Face recognition using Hidden Markov Models , 1995 .

[19]  Anil K. Jain,et al.  Unsupervised Learning of Finite Mixture Models , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[20]  Michael J. Fischer,et al.  The String-to-String Correction Problem , 1974, JACM.

[21]  Dewen Hu,et al.  Globally Consistent Reconstruction of Ripped-Up Documents , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[22]  J. MacQueen Some methods for classification and analysis of multivariate observations , 1967 .