基于高光谱成像技术的全麦粉中DON含量预测方法研究
Prediction Method Development for Detection of DON Content in Wholewheat Flour Based on Hyperspectral Imaging Technology
邢常瑞 1赵双玲 2孙钰莹 1张洁 1李光磊 1赵小旭 3袁建 1鞠兴荣1
作者信息
- 1. 南京财经大学食品科学与工程学院;江苏省现代粮食流通与安全协同创新中心;江苏省粮油品质控制及深加工技术重点实验室,南京 210023
- 2. 河南省农业农村科技教育中心,郑州 450008
- 3. 山东美正生物科技有限公司,日照 102100
- 折叠
摘要
DON是小麦中检出率较高、危害较严重的真菌毒素之一,是小麦减产和品质劣变的主要威胁.本研究首先采集全麦粉中DON含量的短波红外高光谱信息,建立全麦粉中DON含量的预测模型,并采用连续投影算法(SPA)进行特征波长的选取,比较了全波长范围和特征波长下建立的偏最小二乘法(PLS)和支持向量机(SVM)模型.结果表明,基于全波长数据建立的全麦粉中DON含量的最优预测模型为SVM模型,对应的预测集决定系数R2P 为0.640,预测集均方根误差(RMSEP)为1 904.43 μg/kg,剩余预测残差(RPD)为2.00.基于特征波长建立的最优预测模型为Autoscale-SVM模型(SPA-Autoscale-SVM),模型对应的 R2P、RMSEP和RPD分别为0.716、1 640.41 μg/kg和2.06.预测值与测量值的回归系数R2为0.744 9,说明基于特征波长的Autoscale-SVM模型能够预测全麦粉中DON含量的变化.为验证所建模型的稳定性,重新挑选小麦样品进行高光谱图像采集,将独立验证集带入所建模型中,验证集的拟合回归系数为0.717 8,说明该模型能对全麦粉中DON含量进行预测.
Abstract
DON is one of the mycotoxins with the highest detection rate and the most serious harm in wheat,is the main threat to wheat yield reduction and quality deterioration.In this study,the short-wave infrared hyperspec-tral information of DON content in wholewheat flour was first collected to establish a prediction model of DON content in whole wheat flour,and the characteristic wavelength was selected by continuous projection algorithm(SPA).The partial least square method(PLS)and support vector machine(SVM)models were established based on the full wavelength range and characteristic wavelength and compared.The results indicated that the optimal prediction model for DON content in whole wheat flour was SVM model based on the whole wavelength range.The corresponding pre-diction set determination coefficient R2P was 0.640,the root mean square error of prediction set(RMSEP)was 1 904.43 μg/kg,and the residual prediction residual(RPD)was 2.00.The optimal prediction model based on the characteristic wavelength was Autoscale-SVM model(SPA-Autoscale-SVM),and the R2P,RMSEP and RPD cor-responding to the model were 0.716,1 640.41 μg/kg and 2.06,respectively.The regression coefficient R2 between the predicted value and the measured value was 0.744 9,indicating that the SPA-Autoscale-SVM model could predict the change of DON content in whole wheat flour.In order to verify the stability of the established model,wheat samples were selected again for hyperspectral image acquisition,and the independent verification set was intro-duced into the established model.The fitting regression coefficient of the verification set was 0.717 8,indicating that the model could predict DON content in whole wheat flour.
关键词
高光谱成像/呕吐毒素/小麦/PLS/SVMKey words
hyperspectral imaging/vomitoxin/wheat/PLS/SVM引用本文复制引用
基金项目
国家重点研发计划(2022YFD2100204)
国家重点研发计划(2022YFD2100202)
江苏省高等学校优势学科建设工程项目(苏政办发[2018]87号)
出版年
2024
NETL
NSTL
万方数据