Boosting Crop Yield and Quality: Deep Learning-Based Multi-Classification of Wheat Eye Spot Disease

Wheat eye spot disease threatens global agricultural productivity and food security. This study offers deep learning (DL) hybrid convolutional neural network (CNN) and recurrent neural network (RNN) model for multi-classifying wheat eye spot illness based on five intensity levels. Accurately classifying disease intensity allows prompt treatments and reduces crop losses. A self-collected dataset of wheat photos with varied eye spot disease intensities taught the classifier. Pre-processing the dataset improved image quality and model compatibility. The DL hybrid model captures spatial and temporal data relationships using CNN and RNN architectures. The suggested approach classifies wheat eye spot disease intensity with 95.47% accuracy in experiments. The model's accuracy helps farmers and agricultural specialists make educated decisions and implement disease management methods. The model also exceeds other state-of-the-art models in accuracy and performance. The comparison shows that the developed model classifies wheat eye spot illness across intensity levels better.