<![CDATA[This determination of melon fruit ripeness is an important factor in ensuring fruit quality in terms of taste, texture, and market value. However, ripeness assessment is still predominantly performed manually and relies on subjective judgement, which may lead to decreased product quality, inefficient distribution processes, and potential economic losses. Therefore, an automated approach for classifying melon ripeness levels is required. This study aims to analyze and compare the performance Support Vector Machine (SVM) and Naïve Bayes algorithms for melon ripeness classification based on digital images using Histogram of Oriented Gradients (HOG) feature extraction method. The dataset used in this study consists of 630 melon images divided into three ripeness classes, 209 unripe, 220 semi ripe, and 201 ripe images. The research process includes image preprocessing, data augmentation, feature extraction, model training, and performance evaluation. Experimental results show that the SVM with a Radial Basis Function (RBF) kernel, using parameter C=10 and the default value, achieves the highest classification accuracy of 94%, while the Naïve Bayes algorithm attains an accuracy of 65%. These results indicate that the SVM algorithm demonstrates superior classification performance compared to Naïve Bayes in determining melon ripeness levels.]]>
Copyrights © 2026
