I need to find an aerial photograph, create a new vector layer, trace several fields in this aerial photograph, and find random coordinates within these polygons. I first tried this with the Google plugin. This, unfortunately, uses the Mercator projection, which I think threw everything off.

What is the easiest way to do it?


First, if you don't have source imagery to work from you'll need to acquire a georeferenced orthophoto. If you're in the US you can try using the NAIP photos. NAIP photos are leaf-on photos, and usually 1 meter resolution.

Next, once you have your imagery to work from set your project's CRS to match the photo's. Then create a shapefile and use the image's CRS as it's CRS. With the image and shapefile using the same projection and coordinate system your digitized features should match up more closely.

After you've digitized your polygons you can re-project if necessary to match the coordinate system you want to use for your points.


If the fields are visible in Google Earth the easiest way would be to trace the polygons in GE and save them as KML files which could then be loaded into QGIS. N.


The Openlayers plugin only has problems with high zoom levels in Google Imagery.

You can try bing imagery with the same plugin. Their resolution has improved a lot recently.

Apart from the bug mentioned above, it is no problem creating a vector layer in another projection and using it together with Mercator-projected image tiles for digitizing. The vector layer itself can have UTM metres or lat/long degrees.

Just make sure that the project CRS is EPSG:3857, as long as the google tiles are on screen.

Once you have finished digitizing, you can delete the Google layer and set the project CRS to that of your vector layer.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.