期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2014
卷号：64
期号：2
出版社：Journal of Theoretical and Applied
摘要：Reflectance spectroscopy (RS) is suitable for the composition analysis of mixed pigment, because the corresponding basic pigment can be determined by calculating the spectrum similarity. However, the spectrum similarity between known pigment and mixed pigment may be not the one between known pigment and the most similar basic pigment, and the known pigment with the largest spectrum similarity may be not the highest proportion of mixed pigment, due to the possible relationship among the basic pigments. Thus, it is easy to affect the accuracy of composition analysis. In this study, a composition analysis method of mixed pigment based on spectrum expression and independent component analysis (ICA) is proposed. First of all, the spectral information of mixed pigment is obtained with spectrometer, and is expressed as the mixed signal. After that, the independent signal (spectrum of basic pigment) is generated with ICA. Then, the type of basic pigment is determined by calculating the spectrum similarity between basic pigment and known pigment. Finally, the percentage of basic pigments is obtained by solving Kubelka-Munk equation system. To validate this method, the 4, 5, 6 simulated spectrums were mixed from four spectrums of Munsell color card, and the experiments were carried out in two aspects. On the one hand, the composition analysis of three groups of original simulated spectrums was executed. On the other hand, each group was mixed with one disturbance spectrum in order to test the anti-interference performance of this method. The separated curve is very similar to the original curve. The average similarity is 96.29%, and the maximum one can reach to 98.37%. The calculated percentage of basic pigment closes to the original one. It can be seen that this method is suitable for composition analysis of mixed pigment.
