Satellites photograph the entire Earth regularly. Remote sensing turns these images into information about land, vegetation, water and buildings.
Data sources
- Copernicus Sentinel-2: free multispectral images with 10 m resolution and frequent revisits – ideal for vegetation and land cover.
- Sentinel-1: radar images that see through clouds, useful for floods and ground movement.
- Landsat: free imagery with a long historical archive.
- Commercial satellites: very high resolution for detailed analysis.
- Drones: centimetre-level detail for smaller areas.
Key techniques
- Spectral indices: NDVI measures vegetation health from red and near-infrared light.
- Classification: mapping land cover such as forest, farmland, water and urban areas, increasingly with machine learning.
- Change detection: comparing images over time to see construction, deforestation or damage.
- Object detection: counting buildings, vehicles or solar panels.
Business applications
- Agriculture: crop health, yield estimation, irrigation planning
- Forestry: damage after storms or pests
- Insurance: damage assessment after floods or storms
- Urban planning: growth, green spaces, heat islands
- Energy: site analysis for solar and wind, monitoring infrastructure
- Environment: water bodies, wetlands, compliance monitoring
Workflow
- Define the question and area of interest
- Select imagery (sensor, dates, resolution)
- Pre-process (cloud masking, corrections)
- Analyse and classify
- Validate with ground truth
- Deliver maps, statistics or dashboards
Tools
Google Earth Engine, Python (Rasterio, GDAL), QGIS, ESA SNAP and machine-learning libraries.
Our team has an academic background in geospatial science and remote sensing – we turn imagery into reliable information for your decisions.