SkyTruth / MTR

Mountain Top Removal
Other
8 stars 2 forks source link

Conduct Accuracy Assessment #76

Closed cjthomas730 closed 8 years ago

cjthomas730 commented 8 years ago

See #68 for background

We are ready to begin the point classification process which we will use for our statistical analysis.

The directions are found in the Point Analysis Process document, please read before you begin.

The sample points are in this spreadsheet.

Record the output of your classifications in this spreadsheet

apericak commented 8 years ago

I've skipped Kappa 2011, 2012, 2013, 2014 due to few if any mining points

JerrilynGoldberg commented 8 years ago

I've skipped Kappa 2002 for the same reason.

cjthomas730 commented 8 years ago

1999 has the same issue

cjthomas730 commented 8 years ago

@apericak it looks like the point classification is very close to finished (~90%). The only remaining areas have been classified as error years. I'll look through the non-kappa sites which still have error classifications and see if I can make some sense of them.

cjthomas730 commented 8 years ago

@apericak I believe I've found a suitable replacement area for kappa

apericak commented 8 years ago

@cjthomas730 Great, whenever you have time you can create the random points as you did before. I can help create the RGB, if you want.

cjthomas730 commented 8 years ago

@apericak I updated the spreadsheets, and removed Kappa from the Output Classifications sheet

cjthomas730 commented 8 years ago

@apericak , here's the final sample area. EPSG:3857 Omega_area_EPSG-3857.zip

EPSG:5072 Omega_area_EPSG-5072.zip

I adapted your script to run just for this area: https://code.earthengine.google.com/4309ec8e4e243fc2325ccc4da0ce20cf