Drone-Based Anomaly and Person Detection System

The rapid adoption of drones across diverse sectors demands intelligent systems capable of real-time anomaly and person detection. This paper presents a study on integrating artificial intelligence techniques, including computer vision and deep learning, to design an autonomous drone system for enhanced situational awareness. The work reviews recent developments in drone-based anomaly detection and person identification, focusing on four domains: sensor fusion, AI-driven detection, identification frameworks, and real-time communication mechanisms. The analysis outlines the effectiveness and challenges of existing models, emphasizing their high detection accuracy under controlled settings and reduced reliability in dynamic environments. To address these limitations, a unified architecture is proposed that combines vision-based anomaly detection, intelligent identification, and efficient data transmission for improved surveillance performance. The study concludes by highlighting future research opportunities such as adaptive multi-environment detection, context-aware analytics, and low-latency real-time operation, contributing to the advancement of reliable and intelligent drone surveillance systems.