This study aims to develop a digital image analysis system that can determine the quality and maturity level of roasted coffee beans, in terms of detecting roasted coffee beans that are suitable and unfit for consumption and sold as specialty coffee as contained in the standard coffee bean classification. provided by SNI No. 01-2907-1999. This study focuses on the physical changes of coffee beans after the roasting process, changes that affect in terms of quality, namely shape. Shape extraction uses Metric because the shape to be searched for is a circle shape, and texture extraction uses Gray Level Co-Occurrence Matric (GLCM). This research began by collecting data in the form of 2D digital images of roasted coffee beans. The system developed in this study consisted of two main stages, namely training and testing. The number of coffee bean image data used is 90 images. In analyzing the quality of coffee beans, the data used is an image of coffee beans which consists of two levels, that is good and bad beans. Classification using Naive Bayes algorithm. Based on the results of research on coffee bean quality analysis, the highest training accuracy was 88% and the highest test accuracy was 90%.
Copyrights © 2023