Klasifikasi Citra Kualitas Bibit dalam Meningkatkan Produksi Kelapa Sawit Menggunakan Metode Convolutional Neural Network (CNN)

Abstract

Palm oil is one of the plantation commodities that has a strategic role in Indonesia's economic development. Lack of employee knowledge about the types of diseases in oil palm seedlings resulted in errors in handling them. In the selection of seeds to be planted in plantations, sick will cause unstable palm growth and even die. This study aims to find the best solution according to experts when the emergence of pests or diseases is identified through the pattern and color of the leaves. The data used in this study comes from image data of PT.Gatipura Mulya which is conducting a nursery as many as 612 images of oil palm seedlings and can be divided into 4 classes. The method that can be used in this identification is Convolutional Neural Network (CNN) which can study objects in image patterns. The result of this research is that the accuracy of image recognition is very good. So that this research can be recommended in the introduction of oil palm image patterns.