This paper reviews recent progress in Smart Surveillance Systems, with an emphasis on their roles in crowd control, crime prevention, and behavior monitoring in educational and workplace environments. The adoption of technologies like Machine Learning and Artificial Intelligence, particularly deep learning models such as You Only Look Once (YOLO), has greatly improved the ability to detect and evaluate events in real-time. The paper also explores the drawbacks of traditional surveillance systems, including human error and inefficiencies, and how modern AI-driven solutions are addressing these issues. In addition, it discusses key approaches in face detection, behavior analysis, and anomaly detection, along with a comparative evaluation of various algorithms in intelligent surveillance. It also highlights potential energy-saving strategies and future developments in AI-driven surveillance.