IoT-Based Soil Nutrition and Plant Disease Detection for Smart Agriculture Using Residual Attention Networks

Abstract

The integration of Internet of Things (IoT) and Artificial Intelligence (AI) has revolutionized modern agriculture by enabling intelligent soil monitoring and early plant disease detection for precision farming applications. Traditional agricultural methods often depend on manual inspection and experience-based decision-making, leading to delayed disease identification, inefficient fertilizer usage, and reduced crop productivity. Recent advancements in deep learning and optimization techniques have introduced automated and data-driven solutions for smart agriculture systems. This survey reviews IoT-based soil nutrition and plant disease detection methods, focusing on advanced architectures such as Convolutional Neural Networks (CNNs), Vision Transformers (ViTs), hybrid CNN-LSTM models, and optimization-driven frameworks. Particular attention is given to the Multi-Layer Stacked Residual Coordinate Boosted Sooty Tern Attention Network, which integrates residual learning, coordinate attention mechanisms, and optimization strategies to improve feature extraction and classification accuracy. IoT devices continuously collect environmental data including soil moisture, nutrient levels, humidity, and temperature, which are analysed using AI models to support intelligent agricultural decisions. Recent studies demonstrate that hybrid deep learning frameworks significantly outperform conventional approaches in disease prediction and fertilizer optimization. However, challenges such as computational complexity, scalability, energy constraints, and large data requirements remain critical issues for real-time IoT deployment in smart agricultural environments.