<![CDATA[An This study aims to develop an automatic classification system for determining the ripeness level of bananas using digital image processing and the Support Vector Machine (SVM) algorithm. Banana ripeness is commonly assessed visually based on skin color, which is subjective and prone to inconsistency. To address this issue, a computer-based classification approach is proposed to improve accuracy and objectivity. The dataset used in this study consists of banana images categorized into three ripeness levels: unripe, ripe, and overripe. The images were obtained from direct acquisition using a smartphone camera and an online dataset platform. The preprocessing stage includes image resizing, color space conversion, and normalization. Feature extraction is performed using color features in the HSV color space combined with texture features extracted using the Histogram of Oriented Gradients (HOG) method. The extracted features are then classified using the Support Vector Machine algorithm with a Radial Basis Function (RBF) kernel. Model performance is evaluated using accuracy, precision, recall, and F1-score metrics. Experimental results show that the proposed SVM-based approach is able to classify banana ripeness levels effectively with satisfactory performance. The results indicate that the integration of digital image processing and SVM has strong potential to support automatic and consistent banana ripeness classification, which can be applied in agricultural and post-harvest quality control systems.]]>
