Planting View – Agri-Environmental Intelligence

Pixels don't tell you about farms.

Pixel-based analysis blurs field edges and mixes crops, making farm-level statistics and agricultural cadastre work inaccurate.

From satellite image to farm database.

Designed for national-scale processing.

• Enhanced analysis at agricultural land and field level
• Transition from pixel-based to object-based processing for improved accuracy
• Support for agricultural cadastre services

• Outlines every field automatically from satellite images, even when neighbouring fields look alike
• Keeps touching fields apart, so each one gets its own clean shape
• Builds a farm database at county, state or country level

• Maps Iran's farmland as irrigated or rainfed at 10 m detail
• Combines Sentinel and Landsat imagery with climate data, so past years can be mapped too
• Adapts to local crops from only a small number of labelled fields

• Follows each field's growth stage, planting date and growth delay
• Flags weeds, water stress and nutrient shortage on the map
• Irrigation advice with a 14-day forecast

• Tracks land use, vegetation, urban growth and infrastructure over time
• Follows changes across the years, such as irrigated wheat area per province
• Helps plan seasons and suggest crops that suit local water, climate and soil

Results.

Scaling for broader impact.