Mosaicking

Optimizes the image coverage search to avoid scene overlaps. Only downloads the neccessary parts.

Download the overlap area, and each unique area in a separate job. For the overlap area the scene with the lowest cloudcover is selected.

%load_ext autoreload
%autoreload 2
%matplotlib inline

from utils.widgets import UI
UI = UI()

Authenticate

UI.authenticate()

Select output folder

UI.choose_output_folder()

Select input AOI

The AOI should be a geojson FeatureCollection with a single feature.

UI.load_aoi()

Search available images

UI.create_search_params()

UI.search_available_images()

UI.show_quicklooks()

Optimize coverage

Get best non-overlapping sections. Iteratively prioritizied by cloud cover & area

UI.optimize_coverage()

Create workflow and run test jobs

Note the mapping between the data sources and the block name:

• Pléiades Reflectance (Download) -> oneatlas-pleiades-fullscene
• Pléiades Display (Download) -> oneatlas-pleiades-display
• Pléiades Display (Streaming) -> oneatlas-pleiades-aoiclipped
• SPOT 6/7 Reflectance (Download) -> oneatlas-spot-fullscene
• SPOT 6/7 Display (Download) -> oneatlas-spot-display
• SPOT 6/7 Display (Streaming) -> oneatlas-spot-aoiclipped

UI.test_workflow()

Run jobs

UI.run_workflow()

Mosaic

UI.mosaic_sections()

UI.view_mosaic()