LandGuard

LandGuard — From satellite intelligence to rapid flood-damage mapping, identifying damaged infrastructure and cut-off communities, even when connectivity fails.

Comment

The Story Behind LandGuard

The Inspiration

During a flood, knowing that an area is affected is only the beginning. Emergency teams also need to know where the flood reached, what infrastructure was affected, and which communities may have become isolated.

When roads, bridges and communication infrastructure are damaged, ground-based information can become incomplete or delayed.

This inspired us to build LandGuard — a space-based disaster intelligence platform designed to transform satellite observations into actionable information for emergency response.

Our core idea is simple:

See the damage. Understand the impact. Reach the people.

What LandGuard Does

LandGuard combines satellite imagery, artificial intelligence, geospatial analysis, road-network analysis and resilient mobile communication into one disaster-response workflow.

The system is designed to take an area of interest and an event date, process satellite and geographic information, and generate an interpretable picture of the disaster.

The workflow uses:

  • Sentinel-1 SAR for before/after radar-based flood analysis.
  • Sentinel-2 optical imagery for complementary surface information when suitable imagery is available.
  • Copernicus DEM for elevation and terrain analysis.
  • Pre-event OpenStreetMap data for buildings, roads and bridges.
  • AI-based segmentation to identify flood-affected areas.
  • Geospatial overlay analysis to estimate potentially affected infrastructure.
  • Road-network analysis to identify settlements that may have lost connectivity to important locations such as towns or hospitals.

The resulting intelligence is presented through an interactive GIS dashboard and the LandGuard Android application.

From Satellite Data to Decisions

Our system follows an end-to-end pipeline:

Area + Event Date
        ↓
Satellite & Geospatial Data
        ↓
Preprocessing
        ↓
AI Flood / Change Detection
        ↓
Flood Extent
        ↓
Infrastructure Impact Analysis
        ↓
Road Network Analysis
        ↓
Potentially Cut-off Settlements
        ↓
Situation Report + GIS Dashboard
        ↓
LandGuard Mobile Application

Built With

  • androidapp
  • artificialintelligence
  • computervision
  • deeplearning
  • disasterresponse
  • flooddetection
  • geopandas
  • geospatial
  • gis
  • kotlin
  • machine-learning
  • openstreetmap
  • python
  • remotesensing
  • satelliteimagery
  • sentinel-1
  • sentinel-2
Share this project:

Updates

posted an update —

LandGuard — From Concept to Disaster Intelligence Platform

We have reached a major milestone in the development of LandGuard.

Our project has evolved from an environmental monitoring concept into a space-based flood damage intelligence platform for Multimodal AI Hackathon 2026 — Track B: Mapping Flood Damage from Space.

The current development focuses on:

  • Sentinel-1 and Sentinel-2 satellite analysis
  • AI-based flood segmentation
  • GIS-based infrastructure impact analysis
  • Road-network analysis for identifying potentially cut-off settlements
  • Interactive disaster intelligence dashboard
  • Android-based emergency information delivery
  • Offline peer-to-peer alert communication

We are now integrating these components into an end-to-end workflow that transforms satellite observations into actionable disaster-response information.

From space-based intelligence to information that can reach people on the ground.

More updates will follow as we continue testing and refining LandGuard.

Log in or sign up for Devpost to join the conversation.

Submission history