New Self-Organizing Maps with Non-Conventional Metrics and Their Applications for Iris Recognition and Automatic Translation

This paper has as aim the design and applications of two self-organizing maps using nonconventional metrics. First approach concerns the Levensthein Self-Organizing Map (LSOM). The LSOM is a SOM that uses a symbolic representation for both the input and also for the weight rows and it is based on the Levensthein metrics. The software implementation of the experimental LSOM model is designed for automatic Romanian-English translation of 1009 words and expressions. The second approach corresponds to the Hamming Self-Organizing Map (HSOM). The HSOM is a SOM that uses binary representation for input and weight vectors and is based on Hamming metric. We have implemented the HSOM for recognition of iris binary templates, and we have evaluated its performances for CASIA iris database with 108 subjects. A recognition score of 99.08% is obtained. Key-Words: Self-organizing map, Levensthein metric, Hamming metric, automatic translation, iris recognition