QUALITY CLASSIFICATION OF GREEN BEAN COFFEE USING CONVOLUTIONAL NEURAL NETWORK METHOD

Abstract

ABSTRACT

AZHAR MUHAMAD. T3118209. QUALITY CLASSIFICATION OF GREEN BEAN COFFEE USING CONVOLUTIONAL NEURAL NETWORK METHOD

Coffee is one of Indonesia's foreign exchange sources and plays an important role in the development of the plantation industry. In the commercial process, a product must have advantages, especially in terms of quality to survive in world market competition. The Convolutional Neural Network (CNN) method is a Deep Learning method that can identify and classify an object in a digital image. The training process is carried out by looking for a model structure that matches the training data and validation data so that overfitting does not occur in the CNN network. The experimental results in this study indicate that the Convolutional Neural Network method can classify the quality of green bean coffee with an accuracy rate of 90%, recall of 92%, precision of 86%, and F1-Score of 88% from 30 images by taking 15 sample images from each class using confusion matrix testing.