<![CDATA[The identification of tomato ripeness is essential in agriculture to ensure quality and reduce spoilage. Traditional methods that rely on human observation are often inaccurate. This research aims to develop a CNN-based system to accurately identify tomato ripeness, focusing on the ripe and unripe categories. The images were taken while the tomatoes were still on the plant stalk with data collection involving 1.000 images of tomatoes, obtained from Kaggle andtaking pictures of tomatoesat the Poktan Welan Asri Petinggen garden tour site, Yogyakarta. The dataset was divided into training (700 images), testing (150 images), and validation (150 images). The experimental designuses the CNN model with image processing steps such as resizing, labeling, and data augmentation. Testing on this system achieved an accuracy of 92.67%. These findings demonstrate the effectiveness of CNN in detecting tomato ripeness, providing a reliable solution for farmers and agricultural stakeholders]]>
Copyrights © 2025
