Construction of inner quality estimation model of boiled rice and its application for the cooking process

2016 Fiscal Year Final Research Report

• Project/Area Number: 26350093
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Eating habits
• Research Institution: Kochi University
• Principal Investigator: MURAI Masauki 高知大学, 教育研究部自然科学系農学部門, 教授 (00166240)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: 品質推定 / 炊飯米 / ニューラルネット

Outline of Final Research Achievements

For optimizing cooking condition of rice grains, a neural network estimation model, that relates cooking process patterns with the cooked rice quality, was proposed. In construction of neural network model for estimating quality index, hardness of boil-cooked rice samples, water adsorption and DSC curve patterns were used for inputs of the model. After training the neural network model by back-propagation learning method, the accuracy of the estimated values to the measured data not used for the training was checked. As the result, it was found that relative error of the estimated values to the measured was within 20%.

Free Research Field

育種学