Geospatial Machine Learning at Scale
summary
Amazon SageMaker adds geospatial connectors for Sentinel‑2, Landsat and OpenStreetMap, a notebook SDK to launch managed processing jobs (demo job returns 25 scenes and typically completes in ~20 minutes), built‑in ops like resampling and geomosaic, a cloud removal model that masks and fills clouds, and ready transformations to compute NDVI time‑lapse stored to S3. The demo runs an NDVI pipeline over a Brazil AOI, exports a time series and area measurements, and produces a clear time‑lapse showing steady loss of greenness over several years.
AI-written summary · about
chapters (7)
- 0:00Geospatial ML introduction
- 3:26Geospatial data challenges
- 8:44SageMaker geospatial features
- 13:03Demo: Raw materials monitoring
- 20:31Deforestation analysis with NDVI
- 21:56Resources for geospatial ML
- 23:21Q&A
related talks
48mAccelerate science, research, and innovation with AWS
26mEarthquake damage mapping with Amazon SageMaker geospatial
19mUsing Amazon SageMaker to develop and deploy ML models
31mFrom Radar Echoes To Real-Time Recognition: Deep Learning on Another Planet -An Open Source Workflow
01h 36mFrom Maps to Models: An End-to-End Journey in Geospatial Machine Learning with Python
21mAWS On Air ft. Amazon SageMaker Ground Truth Plus
37mStreamlining Machine Learning Operations
17mFree satellite imagery and DEM analyses enabling NRM in the developing world