Solving weighted graph matching problem by modified microgenetic algorithm
Chengwen Liu, Kuo‐Chin Fan, Jorng‐Tzong Horng, Yuan-Kai Wang · 2002
Microgenetic algorithm (MGA) is genetic algorithm (GA) using a very small population size (population size<20). The weighted graph matching problem (WGMP) has received much attention in the field of pattern recognition. In this paper, a hybrid MGA with larger population is proposed to solved the weighted graph matching problem. In our hybrid microgenetic algorithm, many modules, such as local search algorithm, biased initial population, a modified selection scheme, and a refining procedure, are embedded to improve the performance of the algorithm. Experimental results show that our method outperforms a well-known method, the symmetric polynomial transform (SPT), on most instances of the weighted graph matching problems.