AIoT-Enabled Precision Agriculture for Sustainable Crop Disease Management: Advancing SDGs Through Graph Attention Neural Networks

By 2050, agricultural productivity must increase by 70% if food security is to be achieved throughout the world despite population expansion. For high-risk crops like wheat, rice, and maize in particular, crop diseases provide tremendous obstacles to food security which is a huge challenge towards achieving the Sustainable Development Goals (SDGs). Artificial intelligence (AI), especially deep learning, stands up as a possible answer in this endeavor. This research supports improved disease detection technologies using multi-object techniques based on Artificial Intelligence of Things (AIoT), including Graph Attention Networks (GATs), with an emphasis on wheat disease detection, which is crucial for diabetics globally. The literature review emphasizes the importance of AIoT and GATs in remote sensing, machine learning, gene editing, precision agriculture, and cost-effective crop monitoring systems. The suggested ensemble model, which combines Fast R-CNNs, Mask R-CNNs, and RetinaNet, distinguishes between healthy and Septoria/Stripe rust-infected wheat samples with good accuracy. GATs are useful in identifying patterns in complicated datasets, which improves efficiency. The success of this wheat disease classification technique is demonstrated by experimental validation, which shows an overall accuracy of 92%, precision of 89%, recall of 94%, and F1-score of 91% in disease classification. Early disease detection, enhanced crop management, and lower yield losses are all possible by integrating AI, AIoT, and SDGs-aligned solutions. This work demonstrates the potential of AIoT-driven initiatives to contribute to SDGs objectives for sustainable agriculture and global well-being.