Inspiration
School campuses rely heavily on passive CCTV monitoring. Human-based surveillance is reactive, slow, and error-prone. Rising concerns about unauthorized access, violence, and emergency response delays highlight the need for automated, real-time intelligence. SentinelX was created to shift campus security from manual observation to proactive AI-driven monitoring.
What it does
SentinelX is an AI-powered surveillance system that monitors campus environments in real time. Core functions: Detects unauthorized access and abnormal movement patterns Identifies suspicious behavior using computer vision Sends instant alerts to administrators Displays live analytics through a centralized dashboard The system transforms traditional CCTV feeds into actionable security intelligence.
How we built it
Architecture components: Input Layer: IP camera / webcam feed Processing Layer: Python + OpenCV for video processing AI Layer: Object detection model (e.g., YOLO) for anomaly detection Alert System: Automated trigger (email/notification simulation) Dashboard: Web interface displaying live status and logs Workflow: Camera → Frame Capture → AI Detection → Risk Classification → Alert Trigger → Dashboard Update Technologies used: Python OpenCV YOLO (object detection) Flask (web dashboard) SQLite (event logging database)
Challenges we ran into
Real-time processing latency on limited hardware False positives in motion and anomaly detection Optimizing model performance for small-scale deployment Integrating AI output smoothly into a web dashboard
Accomplishments that we're proud of
Built a functional end-to-end AI surveillance pipeline Achieved near real-time detection with stable frame processing Successfully integrated AI detection with automated alert system Developed a working dashboard demonstrating system usability
What we learned
Real-time AI systems require performance optimization at every layer Detection accuracy depends heavily on dataset quality System architecture design is as important as AI model choice Security solutions must balance sensitivity and false alarm control
What's next for SentinelX – AI Powered Smart Campus Security System
Multi-camera synchronization Cloud-based deployment for scalability Face recognition with access control integration Behavioral anomaly detection using advanced ML models Mobile alert application for administrators Edge AI optimization for low-power devices (Raspberry Pi deployment)
Log in or sign up for Devpost to join the conversation.