Variety Identification And Origin Confirmation Of Songyuan Rice Based On Near Infrared Spectroscopy

The japonica rice produced in Jilin Province has a high status in China.Its unique natural growth conditions make the rice have superior taste and quality,and because its geographical indication rice has great influence in China,it's necessary to study rice origin confirmation technology in this area.In the 21 st century,with economic globalization and the fast circulation of food,and because of the economic profits geographical indication rice has brought,In the production,packaging and circulation of rice,food safety problems such as mixed sales of different varieties and false identification of the real are frequently found.The variety of counterfeit and adulterated Songyuan rice is also circulating in the market,which not only has a bad impact on Songyuan's geographical indication rice,but also has potential safety hazards to consumers.Therefore,it is very important to identify the varieties and confirm the rice of Songyuan origin,and this work has been paid more and more attention by researchers.Relevant studies at home and abroad have shown that near-infrared spectroscopy combined with multivariate statistical analysis is an important,fast,convenient and effective technical means for variety identification and origin confirmation.Most of the existing studies on varieties of agricultural products are from different provinces or across North and South,so is the origin,there are more studies between different provinces,and there are fewer explorations and discoveries for small-scale regional varieties and production areas.Therefore,the near-infrared spectroscopy technique used in this paper combined with PLS-DA(Partial Least Squares Discrimination)to establish the model was designed to provide reference for the later variety identification and origin confirmation.The subject of this thesis is different rice varieties from Songyuan.A total of 368 samples of 5 varieties were collected.The data were collected by Fourier near-infrared spectroscopy,and then the processed data was divided into calibration set and the verification set,the principal component analysis of the data and the establishment of the PLS-DA discriminant was carried out.The validation set is used to verify the model,and the samples from Liuhe and Meihe are used to verify the model.The main conclusions of this study are as follows:(1)Pre-processing and establishing the PLS-DA discriminant model by different methods of spectral data,and determining the optimal pre-processing method for establishing the model.This process shows that the models established based on the data of different pre-processing methods have different effects.The best spectral preprocessing method used in this study is first derivative + Savitzky-Golay9 point smoothing.(2)The PLS-DA discriminant model was established through the correction set data,and the verification result showed that the model could be used for the variety discrimination of different varieties of rice in Songyuan.(3)The Songyuan origin rice was confirmed by using rice samples from Liuhe and Meihe.the rice samples of the same year were treated with the same method,Songyuan,Meihe and Liuhe have 20 samples each and the samples were taken into the Songyuan Dao Huaxiang discriminant model,the results show that this model can distinguish Songyuan rice samples from the other two samples,and it can be used to confirm the rice of Songyuan origin.(4)The results of principal component analysis(PCA)showed that it could discriminate five different rice varieties,however,two of them overlapped partially.The discriminant effect of five varieties tested by PLS-DA discriminant model reached 100%,and the result was better than that of principal component analysis.