RICE QUALITY DETECTION BASED ON DIGITAL IMAGE USING CLASSIFICATION METHOD

Abstract

Rice is one of the staples that is included in the consistent list of staple food commodities (Bapok), currently some irresponsible people make the rice more durable, fragrant and whiter. Many assume that the rice is clean, odorless, and has a high price is rice with good quality and vice versa. From the existing problems the author wants to help the community to better determine good quality rice and good for consumption. This research will create a system that can recognize the type of rice based on the image of the rice. Rice data that has been collected will be sampled and trained using the K-Nearest Neighbors (k-NN) method where this method is used for the classification of the shortest distance calculation which will produce a class in the form of rice data classes, while to obtain parameter values from the rice image using the extraction feature. RGB color average (Red, Green, and Blue) and to get results with a good level of accuracy will use K-Fold Validation.