基于近红外光谱技术对草莓品质的快速检测研究

Rapid Detection of Strawberry Quality Based on Near Infrared Spectroscopy Methods

为实现对草莓内部品质的快速检测,对120份草莓采用近红外漫反射光谱技术结合偏最小二乘法对其品质进行检测研究。结果表明,Vc含量检测模型,其校正集RMSECV为0.761,R^2为97.64%,验证集RMSEP为0.352,R^2为99.52%;可滴定酸度含量检测模型,其校正集RMSECV为0.019 5,R^2为96.81%,验证集的RMSEP为0.013 4,R^2为98.77%;可溶性固形物含量检测模型,其校正集RMSECV为0.713,R^2为95.07%,验证集的RMSEP为0.037 9,R^2为98.75%;硬度含量检测模型,其校正集RMSECV为144,R^2为97.84%,验证集的RMSEP为144,R^2为98.09%。模型的稳定性及检测精度较高,因此采用近红外光谱技术可以实现对草莓品质的快速检测。

In order to realize the rapid detection of the internal quality of strawberries,the quality of 120 strawberries was detected by near infrared diffuse reflection spectroscopy combined with partial least square method.The results showed that the vitamin C content detection model had a calibration set root mean square error of cross validation was 0.761,R2 was 97.64%,a validation set root mean square error of prediction was 0.352,and R2 was 99.52%;the titratable acid content detection model had a calibration set root mean square error of cross validation was 0.0195,R2 was 96.81%,a validation set of root mean square error of prediction was 0.0134,and R2 was 98.77%;the soluble solid content detection model had a calibration set of root mean square error of cross validation was 0.713,R2 was 95.07%,and a validation set of root mean square error of prediction was 0.0379 and R2 was 98.75%;the hardness content detection model had a calibration set root mean square error of cross validation was 144,R2 was 97.84%,a verification set of root mean square error of prediction was 144,and R2 was 98.09%.The stability of the model and the detection accuracy were high,so the near infrared spectroscopy could be used to detect strawberry quality quickly.