支持向量回归算法在NIR光谱法预测烟草淀粉中的应用  

Application of SVR in NIR Prediction Model for Starch in Tobacco

作　　者：张朝平 刘太昂[2] 葛炯 张建平 刘旭[2] 王维妙 孙文梁 束茹欣 杨凯 

机构地区：[1]上海烟草(集团)公司技术中心,上海市杨浦区长阳路717号200082 [2]上海大学理学院化学系,上海市上大路99号200444

出　　处：《烟草科技》2009年第10期41-44,49,共5页Tobacco Science & Technology

摘　　要：为考察支持向量机回归(SVR)在烟草近红外光谱(NIRS)分析中应用的可行性,采用偏最小二乘回归(PLS)、多元线性回归(MLR)、误差反向传播人工神经网络(BP-ANN)和SVR对187份烟草样品的NIR漫反射光谱及其淀粉含量的化学测定数据进行处理,建立了烟草中淀粉含量NIRS定标模型,并采用留一法交叉验证(LOOCV)和独立样本集对模型进行了内部和外部验证。结果表明,SVR模型的预测能力比BP-ANN、PLS和MLR模型略好。因此,可将SVR引入到烟草淀粉含量的NIR分析中。In order to test the applicability of support vector regression （SVR） in near-infrared （NIR） spectral analysis of tobacco, NIR calibration model for starch in tobacco were developed by processing NIR spectra and chemically determined starch content data of 187 tobacco samples with SVR, partial least square regression （PLS） , muhiplicative linear regression （MLR） and error back propagation artificial neural network （BP-ANN）. The obtained model was tested through internal validation and external validation with leave-one- out cross valida model was sligh applied to NIR tion （LOOCV） and independent sample set. The results showed that the accuracy of SVR tly higher than that of BP-ANN, PLS and MLR models, it implied that SVR algorithm could be analysis of starch in tobacco.

关 键 词：烟草 近红外光谱 支持向量回归 机器学习 

分 类 号：TS411.1[农业科学—烟草工业]