CLASSIFICATION OF GUAVA FRUIT QUALITY BASED ON DIGITAL IMAGES USING MACHINE LEARNING

Abstract

Guava fruit quality assessment is an important process to ensure the quality of fruit consumed by the public. However, manual quality assessment is still subjective and time-consuming. Therefore, this study aims to classify guava fruit quality based on digital images using computer vision and machine learning approaches. The guava image dataset is divided into two classes, namely good guava and bad guava. Feature extraction is performed by calculating the average values of the RGB (Red, Green, and Blue) color channels from each image. The supervised learning methods used are K-Nearest Neighbor (KNN) and Support Vector Machine (SVM), while the unsupervised learning method uses K-Means Clustering. The system implementation is carried out using RStudio software. The experimental results show that the KNN algorithm achieves an accuracy of 92.39%, while the SVM algorithm achieves an accuracy of 89.17%. Based on these results, it can be concluded that the proposed method is capable of classifying guava fruit quality effectively and has the potential to be used as an automated classification system.