ABSTRACT

Watermelon is a valuable fruit with a long storage period, and its ripeness is crucial for its taste, quality, and preservation. Traditional methods of assessing ripeness are time-consuming and prone to errors. This study used CNN models to categorise watermelon ripeness into unripe, ripe, and overripe, using Jupyter and several libraries. The MobileNetV2 network performed the best, with an accuracy rate of 95.97%.