利用可见/近红外光谱判别干枣品种

目的利用可见/近红外反射光谱技术快速判别干枣的品种。方法使用光谱仪获取山西永和枣、山西板枣和新疆和田枣3种干枣在345~1100 nm波段范围内的漫反射光谱;分别使用多元散射校正(MSC)法和一阶导数法(1~(st)-D)和二阶导数法(2~(st)-D)对反射光谱进行预处理;对预处理光谱进行主成分分析,全交差验证法确定最佳主成分数量,提取主成分,结合马氏距离法和线性判别法建立品种判别模型,建立模型过程中使用全交叉验证法确定最佳主成分数,将模型应用于干枣的品种判别。结果可见/近红外反射光谱经过MSC处理后提取主成分建立品种预测模型对枣的品种判别结果最好,利用前4个主成分结合马氏距离法建立的判别模型和利用前5个主成分结合线性判别法建立判别模型,对于3个品种的枣的校正和验证判别准确率都达到了100%。结论可见/近红外反射光谱技术可以较好地判别干枣品种,本研究可为可见/近红外光谱技术在于枣品种和产地的快速鉴别和溯源中的应用提供一定的技术基础。

Discrimination of dry jujube cultivar varieties using visual/near infrared reflectance spectroscopy

Objective To discriminate different dry jujube cultivated varieties non-destructively by visual/near infrared (VIS/NIR) reflectance spectroscopy technique. Methods The VIS/NIR reflectance spectrums were collected at 345~1100 nm from 3 kinds of dry jujube, including Yonghe jujube, Banzao jujube and Hetian Jujube. The spectrums were pretreated by multiplicative scattering correction (MSC), first derivative (1st-D) and second derivative (2nd-D) methods, respectively. And then, principal component analysis were performed, and the discrimination model was constructed using principal component based on Mahalanobis distance and linear distance respectively. The cross validation method was used to determinate the optimal number of principal components for constructing discrimination models. Results The discrimination model based on Mahalanobis distance using 4 optimal principal components, which were pretreated by MSC spectrum and the discrimination model based on linear distance using 5 optimal principal components which were pretreated by MSC spectrum had the best results. The 2 models gave good discrimination of 3 kinds of jujube with the accuracy rates of 100%. Conclusion The VIS/NIR reflectance spectroscopy technique is useful for discrimination of different kinds of jujube, which can provide an efficient means for the rapid and nondestructive determination of dry quality.