AI DRIVEN REAL TIME SURVEILLANCE SYSTEM FOR PUBLIC SAFETY

The AI Driven real time surveillance system for public safety utilizing YOLOv8 is an innovative approach designed to enhance road safety by detecting Accident severityes in real-time and promptly alerting relevant authorities. Leveraging advanced deep learning techniques, specifically the YOLOv8 object detection algorithm, this system aims to provide a reliable and efficient method for identifying crash scenarios and initiating emergency responses. The dataset for this project, sourced from Roboflow, comprises diverse images of Accident severityes and normal driving conditions, ensuring robust training and evaluation of the model. Implemented in Jupyter Notebook and deployed using Flask, the system processes live video feeds from dashcams or CCTV cameras, making it a practical solution for real-world applications. The model achieved a remarkable 95% accuracy, highlighting its effectiveness in detecting Accident severityes with minimal false positives and negatives. This project not only contributes to reducing response times in emergencies but also offers potential for integration with emergency services and mobile applications, further broadening its impact. Future work includes enhancing the alert system with real- time location data and continuously updating the model with new data to maintain high detection accuracy.