In this work we examine the applicability of an evolutionary algorithm to the problem of optical character recognition. Classification technique is template matching and minimum weighted error. This kind of problem can be turned into an optimisation problem. In particular, we concentrate segmentation and classification on Genocop III algorithm, proposed by Michalewicz [1] for numerical optimisation for constrained problems using a multi-objective function. The proposed algorithm shows good performances and accuracy.
A Character Recognition Handling Constraints Genetic Algorithm: The License Plate Case Study / Bevilacqua, V.; Mastronardi, G.; Colaninno, A.. - STAMPA. - (2004), pp. 109-113. (Intervento presentato al convegno IASTED International Conference on Artificial Intelligence and Applications, AIA-2004 tenutosi a Innsbruck, Austria nel February 16-18, 2004).
A Character Recognition Handling Constraints Genetic Algorithm: The License Plate Case Study
V. Bevilacqua;G. Mastronardi;
2004
Abstract
In this work we examine the applicability of an evolutionary algorithm to the problem of optical character recognition. Classification technique is template matching and minimum weighted error. This kind of problem can be turned into an optimisation problem. In particular, we concentrate segmentation and classification on Genocop III algorithm, proposed by Michalewicz [1] for numerical optimisation for constrained problems using a multi-objective function. The proposed algorithm shows good performances and accuracy.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
