Estimation of Distribution Algorithms: A New Evolutionary Computation Approach for Graph Matching Problems

The interest of graph matching techniques in the pattern recognition field is increasing due to the versatility of representing knowledge in the form of graphs. However, the size of the graphs as well as the number of attributes they contain can be too high for optimization algorithms. This happens for instance in image recognition, where structures of an image to be recognized need to be matched with a model defined as a graph. In order to face this complexity problem, graph matching can be regarded as a combinatorial optimization problem with constraints and it therefore it can be solved with evolutionary computation techniques such as Genetic Algorithms (GAs) and Estimation Distribution Algorithms (EDAs). This work proposes the use of EDAs, both in the discrete and continuous domains, in order to solve the graph matching problem. As an example, a particular inexact graph matching problem applied to recognition of brain structures is shown. This paper compares the performance of these two paradigms for their use in graph matching.