Overview

The AI Segmentation plugin helps quickly extract vector boundaries from remote sensing imagery for crop monitoring, irrigation planning, and land use management.

Common Applications

Crop Field Delineation

Segment individual crop fields or plots from aerial and satellite imagery. This is essential for:

Precision agriculture planning
Yield estimation by parcel
Subsidy and compliance reporting (CAP, USDA)
Farm management system input

Vegetation Health Monitoring

Segment areas of poor vegetation health (stressed crops, disease patches, drought damage) from drone imagery to plan targeted interventions.

If you have multispectral imagery, load the false-color composite (NIR-R-G) to better visualize vegetation health before segmenting.

Irrigation and Water Body Mapping

Segment irrigation canals, ponds, reservoirs, and other water features from imagery. This is useful for:

Water resource inventory
Irrigation planning and optimization
Flood risk assessment

Land Use Classification

Create sample polygons for different land use types (cropland, forest, urban, pasture, water) to train classification models or for direct mapping.

Typical Data Sources

Source	Resolution	Best For
Drone orthomosaics (RGB)	2-10 cm	Field-level detail, individual plants
Drone multispectral	2-10 cm	Vegetation indices, health assessment
Sentinel-2 satellite	10 m	Regional crop monitoring, NDVI
National orthophotos	20-50 cm	Parcel boundaries, land use

Workflow Example: Field Boundary Mapping

Load your aerial imagery or satellite basemap in QGIS
Add an online layer (XYZ tiles) if you need additional context
Start AI Segmentation and navigate to your area of interest
Click on a crop field to segment its boundary
Use Expand/Contract to fine-tune the edge if needed
Press S to save, then move to the next field
When finished, press Enter to export all field boundaries
Open the attribute table and add a "crop_type" column to label each field

Agriculture & Vegetation Analysis