In the modern era of rapid urbanization and popula-tion growth, the availability of secure and efficient crowd manage-ment systems is crucial for public safety. However, the lack of an automated and structured monitoring system often leads to severe consequences such as stampedes, security breaches, and delayed emergency responses, especially during large-scale public events. This paper introduces DensifyAI, a secure and centralized smart crowd monitoring platform designed to address these challenges. The system implements advanced computer vision and deep learning techniques, specifically the YOLOv8 architecture and density map estimation, enabling real-time detection and spatial analysis of crowds. The platform is built using Python, OpenCV, and PyTorch for backend AI processing, React for frontend visu-alization, and a hybrid database system (MySQL and MongoDB) for efficient data management. The proposed system significantly reduces latency in identifying dangerous crowd surges, minimizes human error, and enhances overall public safety. Results demon-strate that DensifyAI improves surveillance efficiency, reduces the need for massive manual security deployment, and provides a scalable solution for smart cities and educational institutions.