Real time notification system for wildlife movement monitoring in TTD pilgrim routes

The project presents a web-based system to monitor wildlife movement along TTD pilgrim routes using live camera feeds. It employs YOLOv9s, a deep learning model, to detect six classes: lion, tiger, elephant, fox, bear, and human. The system processes video frames, identifies animals, and displays instant alerts on a dashboard. Built with Flask and SQLite, it includes user authentication, live detection, and an analytics module. Users can register, log in, view real-time video with overlaid bounding boxes, receive alerts for animal presence, and analyze detection trends through charts and logs. The dataset from Roboflow ensures robust training across varied conditions. The system logs every detection with timestamp and confidence score, enabling pattern analysis. It supports multiple users and maintains session security. The implementation achieves high detection accuracy and low processing delay, making it suitable for deployment in sensitive zones. The modular design allows easy updates and scalability. This solution automates monitoring, reduces manual effort, and supports safety through timely alerts.

Keywords: Wildlife Monitoring, YOLOv9s, Object Detection, Flask, SQLite, Live Alert, TTD Routes, Dashboard, Analytics, Deep Learning