Jens
faeb58ef6d
GeoIntel release gates / Compile, test, contracts and builds (push) Successful in 1m49s
GeoIntel release gates / Python and npm vulnerability policy (push) Successful in 21s
GeoIntel release gates / Production AI image, SBOM and container scan (push) Successful in 5m39s
GeoIntel release gates / Deploy exact gated revision to Unraid (push) Failing after 58m43s
1.5 KiB
1.5 KiB
Domain Model
Project
A project represents a GeoAI investigation. It is not just a folder. It defines the purpose, spatial areas, datasets, analysis runs, and outputs.
Fields:
- id
- name
- description
- region
- status
- created_at
- updated_at
Relationships:
- has many areas
- has many datasets
- has many analysis_runs
- has many exports
Area
An area is a selected geometry, usually a polygon. It may represent a municipality, manually drawn polygon, bounding box, or uploaded GeoJSON geometry.
Rules:
- Must have valid geometry.
- Must have a calculated area in square meters.
- Must keep original input geometry if uploaded.
- Must be convertible to GeoJSON.
Dataset
A dataset is any input or derived geospatial artifact.
Dataset types:
- raster
- vector
- reference
- detection_output
- segmentation_output
- mask
- tile_set
- export
Dataset statuses:
- registered
- uploaded
- metadata_extracted
- processed
- failed
AnalysisRun
Tracks execution of a pipeline.
Analysis types:
- raster_metadata
- raster_clip
- raster_tile
- vector_import
- object_detection
- segmentation
- qaqc
- change_detection
- export
Detection
A model-detected object. V1 supports bounding boxes converted into geospatial polygons.
Important fields:
- class_name
- confidence
- geometry
- bbox_pixel
- bbox_geo
- source_tile
- model_id
QualityCheck
A measured comparison between predicted features and reference features.
Core metrics:
- precision
- recall
- f1
- mean_iou
- matched_count
- false_positive_count
- false_negative_count