MACHINE LEARNING MODEL FOR CLASSIFICATION OF RICE VARIETIES USING HYBRID FEATURE SELECTION SCHEME

Abstract

This research presents a classification, artificial intelligence base rice. The design and achievement of artificial neural network system that extracts specific shape and texture features from rice classification. Rice of three kinds is presented in this research. Modules of significant of rice image features are identified using a texture feature selection technique and neural network. We have three categories and each category have been samples and we have 90% train and 10% test disjoint data sets and then classify. The future technique inherit range and direction invariance through difference near the reflection data sets as well as it be able to manage effectively even through rice. Sample to be distorted suitable toward quit before owing to a integer of whole drill in them. A considerable a extremely towering organization relation of 90% toward 90.1% be achieve, still with for the classification misshapen rice grains.