<![CDATA[The determination of avocado ripeness is still commonly performed manually, which is subjective and often inaccurate due to its reliance on human visual perception. Traditional methods, such as pressing the fruit surface, may damage avocado quality and are inefficient for large-scale distribution and marketing. This study aims to automatically classify avocado ripeness levels using a deep learning approach based on transfer learning. The proposed method employs transfer learning using Convolutional Neural Network architectures, namely VGG16 and VGG19, which have been pre-trained on the ImageNet dataset. The research stages include image pre-processing such as resizing, normalization, and data augmentation to enhance input quality. Subsequently, model training and testing are conducted by comparing the performance of both architectures using evaluation metrics. The dataset used in this study is obtained from the Kaggle platform and consists of avocado images with various ripeness levels. Experimental results indicate that both models are capable of classifying avocado ripeness with high accuracy, precision, recall, and F1-score, with the VGG19 model achieving the best performance. These findings demonstrate that the deep learning approach effectively addresses the subjectivity and inaccuracy associated with manual avocado ripeness determination. This study contributes to the development of an accurate, objective, and practical image-based avocado ripeness classification system with potential applications in agriculture and fruit distribution]]>
