Sentinel Guard

Sentinel Guard Inspiration : Road accidents and traffic congestion continue to be major challenges in modern cities, often leading to delayed emergency response and loss of lives. Existing surveillance systems rely heavily on manual monitoring, making it difficult to identify incidents in real time. We wanted to build an AI-powered solution that transforms ordinary CCTV cameras into intelligent traffic monitoring systems capable of detecting vehicles, monitoring crowd density, identifying accidents, and providing instant alerts to improve road safety.

What it does : Sentinel Guard is an AI-powered smart traffic surveillance system that analyzes live video streams using computer vision. It automatically detects vehicles and pedestrians, estimates traffic density, monitors crowd movement, identifies potential accidents, and displays real-time analytics on an interactive dashboard.

Key Features : Real-time vehicle detection Crowd detection and counting Traffic density analysis AI-based accident detection

How we built it : We built Sentinel Guard using a modern AI and web development stack. Frontend HTML5 CSS3 JavaScript Backend Flask (Python) AI & Computer Vision YOLOv8 OpenCV Tools VS Code Git & GitHub

The frontend communicates with the Flask backend through REST APIs. Live video frames are processed using YOLOv8 for object detection, while OpenCV handles frame processing and visualization. The processed results are streamed back to the dashboard in real time

Challenges we ran into : Integrating real-time video streaming with the web dashboard. Optimizing YOLO inference for smooth live performance. Synchronizing AI detection results with dashboard analytics. Managing API communication between the frontend and backend. Designing a responsive and user-friendly monitoring interface. Implementing reliable accident detection from live traffic footage.

Accomplishments that we're proud of : Successfully built a real-time AI-powered traffic monitoring system. Developed an interactive dashboard with live analytics. Integrated YOLOv8 for accurate vehicle and pedestrian detection. Implemented real-time traffic monitoring and alert generation. Created a scalable architecture suitable for smart city applications.

What we learned : This project helped us gain practical experience in: Computer Vision using YOLOv8 Real-time video processing with OpenCV Backend development using Flask API integration between frontend and backend

What's next for Sentinel Guard : We plan to extend Sentinel Guard with advanced smart city capabilities, including:

-> Automatic ambulance and emergency service notification -> Police control room integration -> Google Maps traffic visualization -> Automatic Number Plate Recognition (ANPR) -> Cloud deployment for city-wide monitoring -> Mobile application for remote monitoring