<![CDATA[Bananas are one of the most popular fruits consumed worldwide, valued for their nutritional benefits and versatility in various dishes. However, ensuring banana quality, including ripeness and integrity, remains crucial in meeting consumer expectations and maintaining supply chain standards. Manual classification of banana quality can be tedious, prompting the need for efficient methods. In this study, we explore the classification of banana quality using Convolutional Neural Network (CNN) with VGG16 architecture and image augmentation. Leveraging previous research and considering the superior performance of VGG16, we gathered data from Kaggle and evaluated our model's accuracy. The implementation yielded promising results, achieving a peak accuracy of 97.50% with 15 epochs and an 80%-20% training-validation data split. This surpasses previous methods, indicating the effectiveness of CNN with VGG16 in banana quality classification. Keywords: Banana quality, Convolutional Neural Network, VGG16 architecture, Image augmentation, Classification accuracy]]>
Copyrights © 2024
