Enhancing Apple Crop Health: A CNN Model for Disease Identification

Maintaining good orchards and guaranteeing maximum crop yields depend on early recognition of apple leaf diseases. We offer in this work a CNN model intended for detection and identification of apple leaf diseases. With 389 photos divided into four different classes. The study makes use of an open-source Kaggle platform dataset. Effective training of the CNN model—which uses a sequential architecture with 2,579,928 trainable parameters—is made possible by the diversity and balance within the dataset. Strong performance of the model was shown by means of thorough training and validation procedures, with an overall classification accuracy of 89%. Extensive study of the loss and accuracy graphs during training reveals the capacity of the model to broadly apply to fresh, un-processable data. By means of a dependable instrument for early disease identification, this work makes a significant addition to the field of agricultural technology, thereby helping to lower crop loss and support sustainable farming methods.