PostGIS in Agribotics
summary
Australia has ~136,000 farms feeding ~80M people and drone imagery aims to cut waste by adding actionable maps. Pipeline ingests mission JSON and GPX (ogr2ogr), stores flight tracks in PostGIS, creates orthomosaics with OpenDroneMap, computes NDVI from near‑IR and red bands using ST_MapAlgebra on rasters, tiles the result into 200x200 tiles, runs ST_ValueCount/ST_Reclass to flag tiles with >75% “bad” pixels. A $100 modified camera captures NIR; 1,200 photos take ~1.5 days to process but the workflow produces searchable, auditable maps and identified problem areas.
AI-written summary · about
chapters (28)
- 0:00Intro
- 3:16Outline
- 4:12Australia
- 4:52Trade agreements
- 5:54National Farmers Federation blueprint
- 6:30Drones
- 7:45Scout Aerial Media
- 8:57Challenges
- 9:19Farms
- 9:35Perspective
- 10:03Types of Drones
- 10:27Current Capabilities
- 10:55Processing Data
- 12:17Crop Protection
- 12:44Automation
- 13:06Why PostgreSQL
- 14:48Simple Scenario
- 15:43Mission Plan
- 23:56Tracking Data
- 25:00OTR to GI
- 28:22NDVI
- 31:29NDVI History
- 32:21Vegetation Range
- 33:03PostGIS Tools
- 43:55Multispectral Camera
- 44:44Consumer Camera
- 46:32Ortho Photos
- 57:02PostGIS Functions
related talks
39mOpening a treasure trove: The Historical Aerial Photography project
40mGIS capability of PostgreSql using POSTGIS & OSM DATA
15mAustralia to Vanuatu: A visual exploration of the mapping challenges in Asia...
05mLocation Intelligence Down Under...
30mGeospatial analysis made easy with PostGIS and Geoalchemy
25mA Journey Through a Geospatial Data Pipeline: From Raw Coordinates to Actionable Insights
16mMapping the Census with PostGIS, Python, and...
37mIntegration, data sharing, opensource, and the agtech startup ecosystem