LIBRAS Signals Recognition: a study with Learning Vector Quantization and Bit Signature

Learning Vector Quantization is a kind of artificial neural network with a competitive and supervised learning. This technique is commonly used to patterns recognition tasks. This artificial neural network was applied to the LIBRAS signals recognition problem, where the images representation was done with bits signatures. This is an unusual combination form and it seems promising, as inferred by the analysis of the results which are shown in this paper.