Wavelets e uma aplicação na espectrometria

Abstract

Wavelets have played an important role on recent researches. Its applications go from pure math to image processing and many others. Their main property is the representation of functions in a scale space, which allows an efficient observation of the function in an small interval. In this thesis, a mathematical development of the construction of a Wavelet family is presented, as well as an application on spectrometry to obtain petroleum properties. To do so, many others tools are used, such as Fuzzy theory, Genetic Algorithm and clusterization, resulting in a method of properties forecasting. Finally, the method results are presented, using the spectrum of samples provided by LabPetro-UFES.