Design and Development of IoT-Based Domestic Surveillance System with Real-Time Anomaly Detection and Chatbot Integration

Conventional home security systems often rely on costly manual monitoring or basic AI technologies, which lack the ability to provide instant detection of unusual activities or comprehensive real-time surveillance. These limitations can lead to delayed responses, increasing the risk to household safety, particularly for children, pets, and property. This project proposes an advanced loT-based home security system that utilizes the YOLO model, a state-of-the-art deep learning algorithm known for its speed and accuracy in object detection, to address these issues effectively. The system integrates IoT cameras with GPU-accelerated edge devices to process video feeds locally, allowing for efficient detection of anomalies such as unauthorized access, loitering, and movements in restricted zones. The YOLO model's real-time processing capabilities ensure rapid detection and classification of suspicious activities with minimal latency. Alerts are immediately communicated to users via an interactive chatbot interface, enabling seamless and instant updates. In addition, a web-based dashboard offers a user-friendly platform for live video monitoring, system configuration, and historical event tracking. This solution significantly reduces the need for continuous human supervision, improving response times and minimizing errors caused by human fatigue. It is designed to be cost-effective, scalable, and adaptable to meet the diverse security needs of households. By leveraging the advanced capabilities of the YOLO model along with IoT and edge computing technologies, this project provides a modern, proactive approach to ensuring household security, offering peace of mind and enhancing safety for today's smart homes.