【摘要】 目的采用近红外光谱技术（NIRS）建立山药是否硫熏的定性分析模型。方法采集不同硫熏程度山药的NIRS图,采用Discriminant analysis进行相似性判别。通过Constant预处理后建立模型,并进行模型内验证和模型外验证。结果4 000¹0 000 cm^-1山药硫熏情况能够较好地区分,内部验证的准确率达100%,外部验证准确率达到93. 33%。结论采用近红外光谱技术对山药硫熏情况的鉴定是可行的。更多还原

【Abstract】 Objective To establish the qualitative analysis model for sulfur fumigation in Dioscoreaop posita Thunb. using near infrared spectroscopy,so as to establish a rapid method to identify that whether or not the yam was smoked. Methods After collecting the near-infrared spectra of different sulfur fumigating yam. The method of constant was used as pretreatment and discriminant analysis method was used to establish identification models. The model validation and external validation were made,and a sulphur fumigated yam analysis model was established using near-infrared spectroscopy technology. Results In wavelength range of 4 000 10 000 cm^-1,sulphur fumigated yam was able to be distinguished. The accuracy rate of internal validation was 100%and the accuracy rate of external validation was 93. 33%. Conclusion It was feasible to identify sulphur fumigated yam samples by near infrared spectroscopy technique.更多还原