摘要
山楂硬度是衡量果实品质和成熟度的重要指标之一,为实现准确、快速和便捷地检测山楂果实硬度,采用近红外光谱技术结合化学计量学及线性和非线性经典建模方法研究山楂果实硬度的评估预测模型建立。结果表明,支持向量机方法所建模型优于偏最小二乘回归模型,最优模型的RMSEC为0.918,RMSEP为0.895。光谱预测值与山楂硬度具有一定相关性,但预测精度有待进一步提高。
Firmness was one of the important indicators to measure the quality and maturity of hawthorn fruit.In order to achieve accurate,rapid and convenient detection of hawthorn fruit firmness,near-infrared spectroscopy technology combined with stoichiometry and linear and nonlinear classical modeling methods was used to study the establishment of hawthorn fruit firmness evaluation and prediction model.The results showed that the model established by support vector machine was better than the partial least squares regression model,and the RMSEC and RMSEP of the optimal model were 0.918 and 0.895 respectively.The spectral predicted value had certain correlation with hawthorn firmness,but the prediction accuracy needed to be further improved.
作者
刘冬冬
李春花
滕佳鑫
王雪妹
赵志磊
赵昕
LIU Dongdong;LI Chunhua;TENG Jiaxin;WANG Xuemei;ZHAO Zhilei;ZHAO Xin(College of Quality and Technology Supervision,Hebei University,Baoding 071002;National and Local Joint Engineering Research Center for Metrology Instruments and Systems,Baoding 071002;Key Laboratory of Energy Metering and Safety Testing Technology of Hebei Province,Baoding 071002)
出处
《食品工业》
CAS
2022年第8期296-299,共4页
The Food Industry
基金
河北省自然科学基金项目(C2021201011)
河北省大学生科技创新能力培育专项(2021H60311)。
关键词
近红外光谱
山楂
硬度
偏最小二乘模型
near infrared spectroscopy
hawthorn
hardness
partial least squares model
