A deep learning-based system for real-time crowd density analysis and risk detection. Using a multi-column CNN model, it estimates crowd density from images and assists in allocating security and medical resources. Built with PyTorch and Streamlit, it helps prevent overcrowding, manage events, and enhance public safety efficiently.