ResNet-CNN Model for Plant Disease Classification for E-Agriculture Applications

The advancement of an economy's innovation capacity is closely tied to the progress in agriculture. Effective plant disease management is crucial for optimizing crop yields and ensuring agricultural sustainability. This project focuses on developing a Plant Disease Detection and Classification Network (PDDC-Net) by integrating deep learning models. The preprocessing stage standardized the dataset images by eliminating various types of interference. The PDDC-Net employs a Residual Network (ResNet)-Convolution Neural Network (CNN), featuring residual networks, to enhance feature extraction and classification accuracy. The PDDC-Net model demonstrated exceptional performance in detecting and classifying plant leaf diseases. Specifically, it achieved an accuracy rate of 99.84% across various crop diseased leaves, including those from potato, tomato, and pepper plants. Further, the suitable pesticide also suggested for the recognised disease. The proposed PDDC-Net effectively addresses the challenge of plant disease identification with high accuracy, underscoring its potential as a robust tool for agricultural disease management and innovation.