柑桔叶片可溶性糖近红外检测非线性模型研究
为了监督柑桔叶片是否缺乏营养元素,对叶片可溶性糖进行分析。采用近红外光谱技术结合误差反馈神经网络(BPNN)和最小二乘支持向量机(LS-SVM)建立定量剖析非线性模型,运用主成分分析(PCA)进行数据压缩、无信息变量消除算法(UVE)和连续投影算法(SPA)进行有效波段筛选的方法来优化模型的输入变量,提高了模型检测精度。同时,利用Savitzke-Golay平滑(S-G)、多元散色校正(MSC)、导数和基线校正(Baseline)等预处理方法进行数据变换,来确定最佳建模方法。结果表明:波长筛选能优化模型,并提高运算速度,其中PCA优化效果最为明显,可溶性糖的相关系数Rp达到最大为O.91,均方...
Saved in:
Published in | 广东农业科学 Vol. 43; no. 11; pp. 43 - 49 |
---|---|
Main Author | |
Format | Journal Article |
Language | Chinese |
Published |
华东交通大学机电工程学院,江西南昌,330013
2016
|
Subjects | |
Online Access | Get full text |
ISSN | 1004-874X |
DOI | 10.16768/j.issn.1004-874X.2016.11.007 |
Cover
Summary: | 为了监督柑桔叶片是否缺乏营养元素,对叶片可溶性糖进行分析。采用近红外光谱技术结合误差反馈神经网络(BPNN)和最小二乘支持向量机(LS-SVM)建立定量剖析非线性模型,运用主成分分析(PCA)进行数据压缩、无信息变量消除算法(UVE)和连续投影算法(SPA)进行有效波段筛选的方法来优化模型的输入变量,提高了模型检测精度。同时,利用Savitzke-Golay平滑(S-G)、多元散色校正(MSC)、导数和基线校正(Baseline)等预处理方法进行数据变换,来确定最佳建模方法。结果表明:波长筛选能优化模型,并提高运算速度,其中PCA优化效果最为明显,可溶性糖的相关系数Rp达到最大为O.91,均方根误差RMSEP最小为4.82,显著提高了模型的检测精度和稳健性,经过优化的输入变量所建模型,能够满足定量检测的要求,具有一定的可行性。 |
---|---|
Bibliography: | 44-1267/S LIU Yan-de, XIAO Huai-chun, HAN Ru-bing, SUN Xu-dong, ZHU Dan-ning, ZENG Ti-wei, LI Ze-min ( School of Mechatronics Engineering, Eash China Jiaotong University, Nanehang 330013, China ) In order to supervise the nutrional elements of citrus leaves, the soluble sugars in the leaves of citrus were analyzed. Combined with back propagation neural network ( BPNN ) and least squares support vector machine ( LS-SVM ), quantitative analysis of the nonlinear model using near infrared spectroscopy was developed, at the same time, data were compressed using principal component analysis ( PCA ), the effective wavelength bands were screened by Uninformative variable elimination ( UVE ) algorithm and Successive projections algorithm ( SPA ) . These methods were adopted to optimize the input variables of the model, which improved the detection accuracy. And spectra processing methods included Savitzke-Golay smoothing ( S-G ), multiple scatter correction ( MSC ), derivative and baseline correction ( Baseline ) and th |
ISSN: | 1004-874X |
DOI: | 10.16768/j.issn.1004-874X.2016.11.007 |