A Convoluted Learning and Classification Methodology to Predict Wheat Crop Disease Detection Using Artificial Intelligence Logic

Wheat is one of the most vital crops globally, and detecting diseases early is crucial for maintaining crop yield and quality. Traditional methods of identifying wheat diseases are time-consuming and labor-intensive, prompting the need for automated solutions. This study presents a deep learning-based approach for wheat crop disease detection using artificial intelligence (AI) logic, focusing on architectures like ResNet152, VGG19, MobileNetV3, and EfficientNetB0. The dataset comprised images of healthy and diseased wheat plants, with data preprocessing and augmentation applied to enhance the model's robustness. ResNet152 achieved the highest classification accuracy of 98.84%, followed by EfficientNetB0 (95.80%) and VGG19 (96.78%). Data augmentation improved the model's performance across all classes, with class-wise accuracies improving by 3% to 4%. The models were evaluated based on precision, recall, F1-score, and inference time. The results demonstrate that ResNet152 offers the best balance between accuracy and efficiency for wheat disease classification, while MobileNetV3 is optimal for real-time deployment on edge devices due to its lightweight design. This model provides an effective, automated method for detecting wheat diseases, aiding in crop management and disease prevention.