<![CDATA[The palm oil industry plays a crucial role in Indonesia’s economy, making fruit classification by ripeness levels essential to ensuring the quality of palm oil production. This study aims to develop a classification system for oil palm fruits into two categories: ripe and unripe, using a Convolutional Neural Network with the Inception-v4 architecture. The dataset consists of 2,900 images, divided into training (2,000), validation (500), and testing (400) sets. The research stages include data collection, pre-processing (duplicate detection, augmentation, and normalization), model training with Inception-v4, evaluation, and result interpretation. Model performance was evaluated using accuracy, precision, recall, f1-score, and confusion matrix. Results indicate that Inception-v4 achieved the highest validation accuracy of 95% in classifying oil palm fruit. Further experiments were conducted using various optimizers (SGD, Adam, RMSprop, Adagrad, Adadelta) to enhance performance. This study confirms that Inception-v4 is highly effective for oil palm fruit classification and can be applied in plantation industries to improve harvest efficiency and production quality.]]>
Copyrights © 2026
