v1.0-alpha

Features

1. Graph Routing
   • Smart Pathfinding: Leverages Graph Neural Networks (GNN) to predict edge safety and travel costs dynamically.
   • Hybrid Solver: Features a high-performance C++ Routing Engine with a robust Python fallback to calculate optimal rescue paths.
   • Interactive Map: Plan missions, manage nodes, and visualize routes using MapLibre GL JS.
2. Aerial Intelligence
   • Real-Time Detection: Utilizes YOLOv11 to detect survivors, fire, and smoke hazards instantly from drone video feeds.
   • Dual-Mode Surveillance: Supports both Thermal (Person Detection) and RGB (Fire & Smoke Detection) modes.
   • Low-Latency Streaming: WebSocket-based architecture ensures real-time video transmission and processing.
3. Text Analysis
   • Intelligent Classification: Uses an LSTM model to distinguish between Real Disasters and Fake/Absurd reports.
   • Priority Scoring: Automatically assigns priority levels (High, Medium, Low) based on the model's confidence score.
   • Batch Processing: Supports analyzing multiple reports simultaneously from uploaded files (.txt, .csv, .json).