Neural network in rice grading: How Malaysian rice can be graded?

• Main Authors: Che Pa, Noraziah, Yusoff, Nooraini, Ahmad, Nor Hayati
• Format: Conference or Workshop Item
• Language: English
• Published: 2016
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://repo.uum.edu.my/23521/1/ICT4T2016%20252%20256.pdf; http://repo.uum.edu.my/23521/

Rice grading plays an important role in the determination of rice quality and its subsequent price in the market.It is an important process applied in the rice production industry with the purpose ensuring that the rice produced for the market meets the quality requirements of consumer. Two important aspects that need to be considered in determining rice grades; grading technique and factors to be used for grading (usually referred as rice attributes).This article proposes how Malaysian rice can be graded. Twenty one features are proposed to be used.Combination of extensive literature review and series of interview were used in determining the features. A Neural Network (NN) model is proposed to be used with the identified features.For evaluation purpose, expert review has been carried out.The proposed model is believed to be beneficial not only for BERNAS but also to other researchers in the same domain. BERNAS can use the NN model to facilitate their inspection for rice quality.The model can be used as guidance or reference for similar grading works.