PENGUKURAN TINGKAT KEMATANGAN KOPI ARABIKA MENGGUNAKAN ALGORITMA K-NEAREST NEIGHBOR

Abstract

Coffee has an important role in improving the national economy. Coffee is also one of the fourth major export commodities in foreign countries. Assessing the level of ripeness of a good coffee can be seen depending on the type of coffee itself. Arabika coffee will start to ripen on days 310 to 350 and for Arabica coffee types it will start to look ripe at the age of 210 to 250 days. In classifying coffee maturity, the K-Nearest Neighbor (KNN) method can be used. By taking a sample image of 3 arabika coffee grains with different levels of maturity twice. The existing data will be processed using the HSV feature to assess the RGB of the coffee bean image data. Based on the test results that have been determined. An accuracy calculation has been used to measure KNN and HSV's performance in determining the ripeness of arabika coffee. The calculation results show the performance of KNN at K=1 is outstanding,, 93.33%.

Keywords Arabica Coffee, ripeness level, K-Nearest Neighbor, HSV, accuracy