SafeLens AI was inspired by a real problem in cities today. Even though cameras are everywhere, most systems only record events and do not help in taking quick action. Because of this, incidents like accidents or fires are not handled fast. So we decided to build a system that doesn’t just watch, but also understands and responds. We created SafeLens AI, a smart city platform that can monitor city activity in real time, detect and manage incidents, assign priority automatically, help authorities respond faster, and provide AI-based insights. We built the frontend using React and Vite as a Command Center dashboard where users can see cameras, incidents, and live data in one place. The backend was developed using FastAPI, which handles data processing and API communication. We also integrated Google Gemini AI to provide intelligent analysis. In addition, we created a simulation engine that generates real-time incidents for demonstration. The system works by loading data from the backend and displaying incidents on the dashboard. The simulation engine generates dynamic events, and each incident is assigned a priority level. If an incident is ignored for too long, it is automatically escalated. Users can also interact with the AI to get insights and analysis. During this project, we learned how to build a full-stack system, integrate AI into real applications, manage real-time data, and design a clean and simple user interface. We faced challenges such as handling real-time simulation, managing multiple components, integrating AI smoothly, and maintaining a responsive UI. In the future, we plan to connect this system with real CCTV cameras, add computer vision for automatic detection, develop a mobile app for field officers, and include predictive analytics. Overall, SafeLens AI is a step towards building smarter and safer cities by combining real-time monitoring, AI analysis, and automated response. This platform has the potential to be used in real-world smart city infrastructure.
Log in or sign up for Devpost to join the conversation.