Improved Weighted Quantum Whale Optimization with vision transformer for intrusion detection, atmospheric monitoring and recommendation in smart agriculture

Abstract

Precision farming enables farmers to make informed decisions regarding fertilization, irrigation, and harvesting by leveraging IoT-enabled sensors that collect real-time data on moisture, temperature, soil nutrients, and other environmental factors. Wireless Sensor Networks (WSNs) in agriculture face challenges such as high energy consumption, security vulnerabilities, and limited real-time data processing capabilities. To address these issues, this paper proposes an Improved Weighted Quantum Whale Optimization (IWQWO) integrated with a Vision Transformer (ViT) for secure and efficient environmental monitoring and intrusion detection in smart agriculture. The IWQWO algorithm combines quantum-inspired techniques with adaptive weighting to optimize node clustering, routing efficiency, and anomaly detection, enhancing energy efficiency and system security. Concurrently, the Vision Transformer captures spatial-temporal relationships in sensor data, ensuring high-precision monitoring, improved intrusion detection, and reduced false alarms. The framework also facilitates resource management, supply-demand prediction, and integration of modern IoT technologies with traditional agricultural practices, including automated irrigation, drone-assisted monitoring, and plant disease detection. Extensive evaluations demonstrate that the proposed IWQWO-ViT model surpasses existing approaches in detection accuracy, cost-effectiveness, and network reliability, offering a robust solution for intelligent, secure, and sustainable agricultural automation.