Study on the Prediction of Rice Noodle Raw Material Index Content by Deep Feature Fusion

Abstract

Rice noodle is a special snack in southern China. With the development of the grain industry and the improvement of living standards, choosing the right raw materials to produce high-quality rice noodles has become one of the problems to be solved at present. Therefore, on the premise of satisfying various characteristics of rice noodles, this paper proposed a deep feature fusion method, which combines with machine learning algorithm to achieve the backward prediction of raw material index content of rice noodles. Deep feature fusion can improve the prediction accuracy by multi-layer weighted feature fusion of rice noodles product index. It realizes feature selection and information extraction of multiple dimensions from the original data and makes the information of the original data play more fully. Experimental results show that the highest ${\mathrm{R}}^{2}$ of the single index of the prediction result can reach 0.987, and the $\mathrm{RMSE}$ of single index only 0.0302. The errors between the predicted value and the actual value of the index of water content, starch content, protein content, swelling force and gelatinization temperature are small, which shows the method has a good prediction effect. It can provide a good reference for the selection of raw materials for the production of high-quality rice noodle.