| 样条变换集成罚函数偏最小二乘方法用于光谱数据重构和定量分析 |
| |
| 引用本文： | 成忠,张立庆.样条变换集成罚函数偏最小二乘方法用于光谱数据重构和定量分析[J].分析化学,2009,37(12). |
| |
| 作者姓名： | 成忠 张立庆 |
| |
| 作者单位： | 浙江科技学院生物与化学工程学院,杭州,310023 |
| |
| 摘 要： | 针对高维小样本光谱数据所显现的函数型数据(Functional data)特性、与性质参数的非线性关系及变量间存有的严重共线性,采用了样条变换集成罚函数偏最小二乘回归新技术.它首先以三次B基样条变换实现非线性光谱数据的线性化重构,随后将重构的新光谱矩阵交由罚函数偏最小二乘法(Penalized PLS)构建其与性质参变量间的校正模型,其中罚函数中的光滑因子由交叉验证优化确定以调控模型的拟合精度.最后,通过小麦样品水分含量的近红外光谱定量分析,结果显示该技术光谱数据重构稳健,去噪明显,并有效解决高维小样本的过拟合和变量间的共线性,而预测集的均方根误差(RMSEP)为0.1808%,方法的非线性校正模型预测能力得到了明显提高.
|
| 关 键 词： | 样条函数 偏最小二乘 粗糙惩罚 近红外光谱 定量分析 小麦 |
Spectral Reconstruction and Quantitative Analysis by B-Spline Transformations and Penalized Partial Least Squares Approach |
| |
| CHENG Zhong,ZHANG Li-Qing.Spectral Reconstruction and Quantitative Analysis by B-Spline Transformations and Penalized Partial Least Squares Approach[J].Chinese Journal of Analytical Chemistry,2009,37(12). |
| |
| Authors: | CHENG Zhong ZHANG Li-Qing |
| |
| Institution: | CHENG Zhong,ZHANG Li-Qing (School of Biological and Chemical Engineering,Zhejiang University of Science and Technology,Hangzhou 310023) |
| |
| Abstract: | Taking into account the near infrared spectra(NIR) on numerous predictor variables with serious collinearity and having nonlinear quantitative relationship with the chemical compositions,a novel nonlinear partial least squares(PLS) approach,termed as Spline-PPLS,was constructed by combining the penalized partial least squares(PPLS) regression with B-splines transformation.Firstly,the observed spectral predictors were considered as discrete observations of curves of the wavelength and were nonlinearly transf... |
| |
| Keywords: | Spline functions partial least square roughness penalty near infrared spectroscopy quantitative analysis wheat samples |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
