Identification of the Mineral Oil Fluorescence Spectroscopy Based on the ICA and SVM

Lv Jiangtao,Gu Zhenpu

DOI: https://doi.org/10.1109/iccda.2010.5540713

2010-01-01

Abstract:Spectral features of the different kinds of the mineral oil are different from each other, so the oil can be identified by the spectral data based on the principle. The composed of it with aromatic hydrocarbons structure is very often complicated, which caused the three-dimensional fluorescence spectroscopy of the different oil are various. The characteristic of the oil style-book are difficult to be maintained by the simple formula when the three-dimensional fluorescence spectroscopy technology is used to identify the species of the mineral oil. In this paper, the independent component analysis (ICA) is used to do the matrix decomposition from the perspective of independence to extract the main feature of the spectroscopy. The support vector machine (SVM) is used to assort the main characteristic root books which are abstracted by the ICA. The species identification of the mineral oil will be realized by it. The identification result is visualized by the parallel coordinate's graph. The experiment results show that it is effective to extract the main feature of the spectroscopy. The classify speed is greatly increased. The discrimination is 98.56. It can effectively realize the identification of the oils.