Empowering Sustainable Farming: Detecting Potato Leaf Diseases with Fine-Tuned VGG19

This is Focussing on late blight, early blight, and healthy leaves, this work properly identified Potato Leaf Diseases using a carefully trained VGG19 model. The dataset consists on remarkably precise photographs of potato plants; preprocessing of these images by means of scaling, standardising, and data augmentation increases model accuracy. First initially Frozen the first layers and replaced a final classification layer to fit the particular work after trained in advance using the ImageNet dataset. By means of this fine-tuning, the model learnt traits specific to Potato Leaf Diseases while preserving the broad feature extraction capacity of the model trained before. An 87% accuracy was obtained by training and testing the model on 3,000 photos. F1-scutes Think back and High accuracy over all classes revealed in the categorisation shows the capacity of the model to correctly recognise several forms of diseases. This method highlights the possibilities of using transfer learning with deep neural networks such VGG19 for agricultural diagnostics, so offering a consistent and automatic approach for identifying potato leaf diseases, thus improving crop management and lowering loss due to plant diseases.