Hi,
I am trying to use the Agriculture Field Delineation pre-trained model on a downloaded Sentinel-2 L2A imagery composite. But I always get an empty output. Can you share the right order & resolution of bands to be used for this workflow? It is unclear which bands are used in the guide shared.
Following is a screenshot of the bands used for creating a composite.
Hi S6,
From looking at the DLPK information, it seems that it requires the following input.
I do not see a specific default order specified. Can you indicate what raster product or band combination you are currently using?
Hi JefferySwain,
The band order is as shown in the screenshot.
Hello S6,
This model expects the band order you have shared. We trained the model using Sentinel-2 Cloud-Optimized GeoTIFFs from Registry of Open Data on AWS: https://registry.opendata.aws/sentinel-2-l2a-cogs/ Since the bands are separate .tif files and of varying spatial resolutions we used a Mosaic Dataset to manage everything. I've just posted a public repository that you can use to automate creation of your own Mosaic Datasets to use with this Deep Learning Model. https://github.com/dkwright/arcgis-sentinel-2-cog-ag-fields/ . There is an example you can run from the batchfiles folder that will create the Mosaic Dataset for San Luis Valley for the month of July in 2023 using a 20% cloud cover limit and it adds a couple dozen processing templates for rendering various imagery composites and band indices on-the-fly. The batchfile can be modified to use your own Area of Interest, acquisition date range, and cloud cover percentage threshold.
I have a problem running the batchfile? I changed the directories but did not do anything
Hi johnpol,
Your python environment path (pPath) should be that of ArcGIS Pro. Default path to this is: "C:\Program Files\ArcGIS\Pro\bin\Python\envs\arcgispro-py3\python.exe" Please verify on your system.
Thank you @DavidWright for sharing this amazing tool. I am able to run this tool, however, I am not getting the raster files. The generated dataset is only giving footprint polygons. The log file is below:
Thanks hafzaal4242!
Please be sure and zoom in to a scale larger than 1:500k as the Mosaic Dataset gets created with some default values for the *PS fields in the footprint table. These field values control the scale range in which to render rasters. When zoomed out to smaller scales it would take a long time to render every image in the Mosaic Dataset.