This study presents a deep learning approach for classifying grapevine leaf images using VGG-16 and VGG-19 neural networks. The research aims to develop an efficient system for grapevine leaf classification, which can be applied in viticulture for disease detection and plant health monitoring. A publicly available dataset consisting of 500 images, categorized into 5 distinct classes (100 images per class), was used for training and testing. The study evaluates the performance of VGG-16 and VGG-19 models in classifying grapevine leaves, with the VGG-19 model achieving a perfect accuracy of 100%, while VGG-16 achieved 99.6%. The results indicate that VGG-19 outperforms VGG-16 in image classification tasks for grapevine leaves. The proposed system includes six stages: loading the dataset, resizing images, loading the deep learning models, training the models, testing images from the internal and external datasets, and evaluating the model's performance. The system's performance is assessed using accuracy, precision, recall, and specificity. The results show that the VGG-19 model provides higher accuracy and better performance in classifying grapevine leaves. The study also compares the proposed method with previous studies, demonstrating its superior performance. The findings suggest that deep learning models, particularly VGG-19, are effective for grapevine leaf classification, offering high accuracy and efficiency in image classification tasks. The study contributes to the field of agricultural applications by demonstrating the potential of deep learning in image classification for grapevine leaves.This study presents a deep learning approach for classifying grapevine leaf images using VGG-16 and VGG-19 neural networks. The research aims to develop an efficient system for grapevine leaf classification, which can be applied in viticulture for disease detection and plant health monitoring. A publicly available dataset consisting of 500 images, categorized into 5 distinct classes (100 images per class), was used for training and testing. The study evaluates the performance of VGG-16 and VGG-19 models in classifying grapevine leaves, with the VGG-19 model achieving a perfect accuracy of 100%, while VGG-16 achieved 99.6%. The results indicate that VGG-19 outperforms VGG-16 in image classification tasks for grapevine leaves. The proposed system includes six stages: loading the dataset, resizing images, loading the deep learning models, training the models, testing images from the internal and external datasets, and evaluating the model's performance. The system's performance is assessed using accuracy, precision, recall, and specificity. The results show that the VGG-19 model provides higher accuracy and better performance in classifying grapevine leaves. The study also compares the proposed method with previous studies, demonstrating its superior performance. The findings suggest that deep learning models, particularly VGG-19, are effective for grapevine leaf classification, offering high accuracy and efficiency in image classification tasks. The study contributes to the field of agricultural applications by demonstrating the potential of deep learning in image classification for grapevine leaves.