YOLOX Driven Smart Surveillance for Real Time Intelligent Object Detection and Anomaly Monitoring

Surveillance systems have been an essential part of the security and situational awareness in public places, industrial zones, and smart cities. Conventional surveillance systems often fail in three significant aspects: accuracy, realtime detection, and adaptation to the situation. This paper presents an advanced YOLOX-based intelligent surveillance system for object identification, categorization, and anomaly detection in real-time. The proposed system utilizes deep learning and computer vision to effectively detect objects, track movements, and trigger an alarm for anomalous activity. The performance of the system is improved by combining edge computing with AI-driven analytics, decreasing latency, and enhancing response time. The results show significant gains in detection accuracy, computing efficiency, and flexibility when compared to traditional YOLO-based models. This research aims to assist in building robust intelligent security systems for practical applications.