Discrimination of Tea by Means of Electronic Tongue and Nonlinear Dimensionality Reduction Model

摘要：

In an electronic tongue system, dimensionality reduction toward the huge data sampled from multisensor array is crucial for the performance of pattern classification. In this paper, Kernel-based nonlinear dimensionality reduction methods were employed to classify different grades of green tea. A comparison of their performances to that of normally used PCA and FLD was presented. Experimental results showed that nonlinear methods could better discover the features that represent the flavor of tea samples. Best discrimination was achieved when Kernel Discriminant Analysis was conducted.

关键词： electronic tongue nonlinear dimensionality reduction kernel discriminant analysis PCA

作者: Ruicong Zhi Lei Zhao Bolin Shi Xingjun Xi

作者单位: Food and Agriculture Standardization Institute China National Institute of Standardization Beijing, Food and Agriculture Standardization Institute China National Institute of Standardization Beijing,

会议类型: 国际会议

会议名称: The 2012 International Conference of Agricultural Engineering and Food Engineering(2012年国际农业工程与食品工程学术会议 ICAE 2012)

会议地点: 哈尔滨

会议语种:英文

页码: 952-955

在线出版日期: 2012-06-16（万方平台首次上网日期，不代表论文的发表时间）