Ensuring safety in public spaces, workplaces, and educational institutions is a critical concern. Traditional surveillance systems lack real-time analysis, making proactive responses to harassment incidents challenging. This project presents an AI-powered surveillance system that detects harassment in real time, fostering a safer environment.
The system utilizes advanced technologies such as Python, OpenCV, MediaPipe, and the YOLO algorithm to analyze live CCTV footage and recognize distress signals and aggressive behavior. Upon detection, it immediately notifies security personnel or authorities, enabling swift action. By integrating with existing CCTV infrastructure, this solution remains cost-effective and scalable.
Key Features: • Real-time video analysis for detecting suspicious activities. • Automated alert system via SMS, email, or mobile apps. • Privacy-focused AI ensuring no personal data is compromised. • Seamless integration with CCTV systems for wide applicability. The AI model is trained on diverse datasets using platforms like Google Colab to ensure high accuracy. The backend, built with Flask or FastAPI, manages real-time processing and alerting. This system transforms surveillance from passive monitoring to active intervention, reducing unreported incidents and improving security.
By leveraging AI for public safety, this project aligns with the goals of the Smart India Hackathon, driving innovation for societal impact.
Log in or sign up for Devpost to join the conversation.