Skip to main content

About the project

πŸ”₯ AGNI-DRISHTI

AI-Assisted Forest Fire Early Warning & Response Intelligence

AGNI-DRISHTI is a geospatial intelligence platform designed to support forest-fire monitoring and response planning in Himachal Pradesh, India. It brings satellite-detected fire hotspots, weather observations, explainable risk assessment, and interactive mapping into a unified operational dashboard.

The platform aims to help users identify areas of interest, understand available environmental conditions, examine hotspot clusters, and explore potential impacted-area scenarios through a clear, map-driven interface.

✨ Key Features

  • πŸ›°οΈ Satellite Hotspot Monitoring β€” Integrates NASA FIRMS observations when the upstream service and valid credentials are available.
  • 🌦️ Weather Intelligence β€” Retrieves available weather observations and forecasts through Open-Meteo.
  • πŸ—ΊοΈ Interactive Geospatial Dashboard β€” Visualizes observations, locations, and geographic context on an interactive map.
  • πŸ€– AI/ML Hotspot Clustering β€” Uses spatial clustering to group nearby satellite detections and highlight concentrations of observations.
  • πŸ”₯ Explainable Risk Assessment β€” Presents a transparent heuristic assessment based on available inputs rather than unsupported prediction claims.
  • πŸŸ₯ Optional Impacted-Area Visualization β€” Provides a switchable map layer for verified fire perimeters, defensible modeled scenarios, or clearly labeled illustrative demonstrations.
  • 🚨 Response Intelligence β€” Organizes observations and presents decision-support information for human review.
  • πŸ”Ž Data Source Transparency β€” Distinguishes live observations, demo data, unavailable services, and degraded integrations.
  • ⚑ Resilient Data Handling β€” Includes input validation, error handling, and appropriate fallbacks to help the dashboard remain usable when services fail.

πŸ› οΈ Technology Stack

  • Frontend: React, TypeScript, Vite, Tailwind CSS
  • Mapping: Leaflet, React Leaflet
  • Backend: Python, FastAPI, Pydantic
  • Data & Machine Learning: pandas, NumPy, scikit-learn
  • Satellite Data: NASA FIRMS
  • Weather Data: Open-Meteo

🧠 AI/ML Approach

AGNI-DRISHTI uses machine-learning techniques such as spatial clustering to organize satellite hotspot observations. Its explainable risk assessment is kept separate from clustering and any future fire-spread model.

A hotspot cluster does not necessarily represent one confirmed fire. A satellite detection does not establish a verified ground fire, and a simulated impacted area is not an observed fire perimeter.

Any future fire-spread predictions require suitable landscape and weather inputs, appropriate model validation, and transparent communication of uncertainty.

🎯 Project Objective

The objective is to make forest-fire intelligence more accessible through an integrated geospatial dashboard that helps users inspect satellite observations, understand available conditions, and support informed human decision-making.

⚠️ Data & Safety Disclaimer

AGNI-DRISHTI is a decision-support prototype, not an official emergency-warning or dispatch system. Data availability depends on upstream services, and satellite observations may be delayed or affected by cloud cover and sensor limitations. Risk assessments and illustrative scenarios must not be interpreted as verified fire perimeters or guaranteed forecasts.

Live, demo, and unavailable data states should always be clearly identified. Operational decisions must be based on verified information and guidance from the relevant authorities.

πŸš€ Project Status

Under active development. Features and data availability depend on integration status, source access, and validation.

Built with

claudeopenai

AI tools

Coded withGemini 3.8 FlashAntigravityGemini 3.5 Flash Lite

Prize Categories Entered

  • Event prize

    Hacktoberfest Hack Day Winners

    Submitted