Numeric Character Recognition System for Chilean License Plates in Semicontrolled Scenarios

The development of a computational tool for the recognition of numerical characters already segmented from Chilean license plates, located in semi-controlled scenarios, is presented. Two algorithms are highlighted: one for the Fine Segmentation of the Characters through the K-Means algorithm and a system of fuzzy logic; and a second one for the recognition through learning of the pseudo-real outline of the characters’ skeletona. The tool has an efficacy of 95 % and a processing time of 0.4 s. © 2017, the Authors.