Deteksi Tingkat Kematangan Buah Alpukat menggunakan Metode YOLOv8

Abstract

Indonesia is one of the countries with a wide variety of plant species. The avocado is an important local fruit from Central America and Mexico that grows in most regions worldwide with tropical and subtropical climates, including Indonesia. Avocado plants first appeared in Indonesia in the 18th century, and one of the largest avocado-producing regions in Indonesia is Aceh. Avocados are one of the most sought-after foods in the world due to their high nutritional value. During the ripening process, avocados do not ripen on the tree and only ripen post-harvest, so they must be harvested in the appropriate physical condition to ensure good fruit quality for consumption. To assist the public, particularly consumers, in detecting the ripeness level of avocados and accurately determining their ripeness, a system has been developed to detect the ripeness level of avocados using the You Only Look Once method version 8. There are four categories detected in this study: unripe, half-ripe, ripe, and rotten. In this study, the total dataset used consists of 928 images, divided into two datasets : 742 training data and 186 testing data. The system can detect the ripeness level of avocados in real time. The method used achieves good accuracy, though not yet perfect, at 94,62%.