Research On Determination Of Nutrient Content In Oat Using Near-infrared Spectroscopy

Oat (Avena sativa L), which is monocotyledonous plants, belongs to the Poaceae (also known as the Gramineae) family. Currently, the production of the oat has been over 22 million tons all over the world. Oats are used for animal feed and as human food because of its nutritional value and functional roles. Oat is an excellent source of dietary fiber, such as ?-glucans (2.2-7.8%) and arabinoxylans. And it also contains protein, B complex vitamins, linoleic acid, minerals and phenolic compounds.Oat has the diverse biological activities, including hypoglycemic, hypolipidemic, antioxidant, enhancing human immunity, and reducing the risk of cardiovascular disease. At present, the traditional methods for measuring oat ingredients need long period, high cost, time-consuming chemical operations and have some trouble in human being and environment. Near-infrared spectroscopy technique has some advantages, such as no pre-treatment, simple operation, no pollution, good repeatability, speediness, and simultaneous analysis of multi-component. Today we can find this technique in many fields, such as food, agriculture or medicine.In this paper, with the oat which were produced all over the country as materials, the near-infrared spectral and the chemometric methods were used to analyze and established the predictive model for detecting the protein, fat,?-glucan, polyphenol content in oat in oat. Specific results were as follows:(1) The method for the near-infrared models which were used to detect protein content from oat was built. The spectrum scattering of SNV, mathematics derivative processing of 2441 and modified partial least square (MPLS) regression were the optimum factors of calibration model. The results indicate that the correlation coefficient of the true value and the prediction value was 0.9543, and the root mean square deviation was 0.1607, which showed that the calibration model had better predictive accuracy.(2) The model of the artificial neural network for determining fat content in oat was established after preprocessing near infrared spectrum data and extracting the spectral characteristics by principle component analysis (PCA). The preprocessing of spectrum scattering was the inverse multiple scatter correction (IMSC) and mathematics processing was 2441(2 is the second derivative processing; 4 is the interval point of the second derivative; 4 is the first smoothing interval point and 1 is no secondary smoothing).The structure of the artificial neural network mode was 2-17-1, which was established after extracting 2 principle component as the characteristic variables of the original information. The correlation coefficient of the true value and the prediction value was 0.962 3, and the root mean square deviation was 1.6072.(3) The model of the artificial neural network for determining ?-glucan content in oat was established after preprocessing the near infrared spectrum data and extracting the spectral characteristics. That the preprocessing of spectrum scattering was the weight multiple scatter correction (WMSC) and mathematics processing was 1441.The structure of the artificial neural network mode was 2-12-1, which was established after extracting 2 principle component as the characteristic variables of the original information. The optimal regression model was obtained with R2=0.9197 and RMSE= 0.02336 in calibration set and R2=0.9206 in prediction set(4) The models which were used to detect polyphenol content from oat were built using and near-infrared spectroscopy technology. The modeling band for the optimum calibration model was 1050nm -1350nm by two-dimensional correlation spectroscopy. The optimum two-dimensional correlation spectral preprocessing combination was SD +SM (second derivatives+second smoothing). The optimal regression model was obtained with R2=0.9623 and RMSCV= 0.03053% in calibration set and R=0.9798 and RMSEV=0.003197 in prediction set.This study can provide a theoretical foundation for detecting the content of the protein, fat, ?-glucan polyphenol content in oat using near-infrared reflectance spectroscopy. By NIRS to determination of nutrient content and distinguishing varieties of oat, is of great significant in improving the level of oat determination, detecting various quality indexes rapidly at the same time and protecting the vital interests for farmers and ensuring national grain security.