Research on Application of Fuzzy Pattern Recognition to Qualitative Analysis of Near infrared Spectroscopy
Multiple linear regression (MLR), the principal components analysis (PCA) and partial least squares (PLS) method are the traditional chemometric methods in the near infrared spectral analysis. However , these linear methods could not obtain the very good predicted accuracy. In this paper, the research on application of Fuzzy Pattern Recognition to qualitative analysis of Near infrared (NIR) spectroscopy is mainly completed. In order to reduce the complexity of analysis model and improve the prediction accuracy, using principal component analysis (PCA), dimension of spectrum variables is reduced. Simultaneously, the crucial questions, closeness degree, principle of choosing the nearest as well as analysis steps, had also solved during the process of analysis. Then the identification model of producing area is established and examined through the prediction samples. The simulation experiment indicates that the prediction accuracy is achieved 97.5％.With the simple modeling process and the stable analysis results, the research has certain application value.
fuzzy pattern recognition Near infrared spectroscopy principal components analysis qualitative analysis
Yong Zhang
Changchun Normal University Changchun, PR. China
国际会议
杭州
英文
481-484
2012-10-28（万方平台首次上网日期，不代表论文的发表时间）