Energy-Efficient Target Coverage in Wireless Visual Sensor Networks Using LSTM-Based Trajectory Prediction and Grey Wolf Optimization

The resource-intensive nature of visual data processing presents significant challenges for Wireless Visual Sensor Networks (WVSNs) in balancing target coverage and energy consumption. The integrated framework presented in this paper combines Long Short-Term Memory (LSTM) networks for trajectory prediction, YOLOv8n(nano)-based object detection, and Grey Wolf Optimization (GWO) for energy-conscious camera node scheduling. The proposed work uses a Pareto-optimal solution space exploration to solve the multi-objective optimization problem of maximizing target coverage while minimizing energy consumption. According to experimental results, our approach reduces energy consumption by 30% compared to baseline methods while achieving up to 95% target coverage. Through chunked inference, the framework effectively processes 10,000 images, produces balanced scheduling decisions across distributed camera nodes, and produces accurate trajectory predictions with MSE loss below 0.01.