mthomp89 / landslide-detect Goto Github PK

Found this while I was researching all of the GEE errors. It replaces the marker we have now with an exclamation point inside a red triangle. I have this in my making_dry notebook.

In the section titled, "Folium Map of All Verified Landslide Locations"
for index, location_info in landslide_gdf.iterrows(): folium.Marker([location_info["lat"], location_info["lon"]], popup=location_info["type"], icon=folium.Icon(color='red', prefix='fa', icon='exclamation-triangle')).add_to(mp)

Note the above is not keeping the correct indentation

@LManak add academic citations

I am wondering if we can create a class or something similar for getting the Bounding Box instead of having to rely of our JSON files. Can we do something where it matches the geometry filter in the image_search function?

Build workflow diagram for end-to-end processing

@LManak Enable json build from dataframe with minimum and maximum coordinates for bounding box

Add AOI center point to dataframe then search by ee method:
point = ee.Geometry.Point(Long, Lat)

I think this is an issue with:
BASE_DATE = ee.Date(landslide_gdf['slide.date'])

the ee.Date seems to have this error: "Invalid argument specified for ee.Date(): slide.id"

We are currently filtering the following, however it could be worth looking into and playing around with different parameters:
.filterBounds(aoi)
.filterDate(ee.Date('2016-09-01'), ee.Date('2017-09-01'))
.filter(ee.Filter.eq('orbitProperties_pass', 'DESCENDING'))
.filter(ee.Filter.eq('relativeOrbitNumber_start', 27))
.filter(ee.Filter.listContains('transmitterReceiverPolarisation',
'VV'))
.filter(ee.Filter.listContains('transmitterReceiverPolarisation',
'VH'))

once we put the new geodataframe from (create_dtaframe) into our code, we need to make sure that it is compatible with everything below. Esp lat long (etc)

is there a formal way to display this information- e.g. relative start number. What other information is in the filename that could be useful?

This is for the iteration workbook. We discussed adding a new column for the center point that is based on the "ge.lat/long" values. This will allow us to work with both the regular lat/long and the ge.lat/long (which is the more accurate verified site).

@mthomp89 We still need to find a way to have this be more reproducible. Should we have the csv of the landslide dataset on Git Hub for future downloads so we don't have to do it manually on our own computer/ so other people have access to the csv when they create their working directory? I don't see it on GitHub yet. I think we just manually added it to our landslide-detect folder.

@LManak office hours (Elsa) orbit number

We discussed making the current dataframe from the DF iteration notebook a geodataframe to allow for more diverse applications

Instead of blue and red, look into what makes those colors, and if the red/blue is confusing for what data we are actually displaying!

This could be a great way to demonstrate the change over time from the landslide. The split plot will show the before/after image side-by-side.

here is the documentation