search

OpenGeoscience / geonotebook

A Jupyter notebook extension for geospatial visualization and analysis
Apache License 2.0
1.08k stars 141 forks source link

Any consideration to integrate Earthengine into geonotebook? #116

Open suredream opened 7 years ago

suredream commented 7 years ago

Hi:

I wonder do you guys want to integrate [Earthengine]() into geonotebook? Or just keep focus on open source gis software?

I would like to contribute in this part if you guys are interested.

Thanks,

aashish24 commented 7 years ago

hi @suredream by earhengine do you mean google earth engine? We could have an extension to support gee. It would be better if we have some mechanism to keep it as plugin but we can figure something out.

suredream commented 7 years ago

Google Earth Engine combines a multi-petabyte catalog of satellite imagery and geospatial datasets with planetary-scale analysis capabilities and makes it available for scientists, researchers, and developers to detect changes, map trends, and quantify differences on the Earth's surface.

​Check it out here:

https://earthengine.google.com/​

dorukozturk commented 7 years ago

Hey @suredream, actually we talked about this in the past both internally and with Earth Engine people. In order to have a meaningful integration with Earth Engine we should be able to figure out a way to collect data from Earth Engine which was kind of not recommended in the past. Our goal in Geonotebook is not just serving tiles but also giving access to the data itself (ideally as numpy arrays). In this case we can use their infrastructure only for serving tiles (slippy style like layer/z/x/y) but getting a subset of a region and exporting that as a numpy array would be very challenging. Do you have a specific use case or do you think the integration will be cool? :)

suredream commented 7 years ago

Doruk Ozturk:

You are right. Get subsetting of the imagery is expensive for earthengine. If you want to get a publish-ready figure, based on earth-engine source and geonotebook visualization feature, you have to go with tiling map.

All the goal is to make spatial map easy. I used to want to write a plugin in QGIS to help composing but QGIS is not a flexible way to do that. So far as I know people still need to export data from GEE to local, and finish map composing in QGIS. Did you face such problem before?

Think about doing a large area mapping, coloring and annotating with current workflow, you will see whether it is worthy to find a way to access EE in geonotebook.

On Fri, Apr 28, 2017 at 1:38 PM, Doruk Ozturk notifications@github.com wrote:

Hey @suredream https://github.com/suredream, actually we talked about this in the past both internally and with Earth Engine people. In order to have a meaningful integration with Earth Engine we should be able to figure out a way to collect data from Earth Engine which was kind of not recommended in the past. Our goal in Geonotebook is not just serving tiles but also giving access to the data itself (ideally as numpy arrays). In this case we can use their infrastructure only for serving tiles (slippy style like layer/z/x/y) but getting a subset of a region and exporting that as a numpy array would be very challenging. Do you have a specific use case or do you think the integration will be cool? :)

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/OpenGeoscience/geonotebook/issues/116#issuecomment-298101036, or mute the thread https://github.com/notifications/unsubscribe-auth/ABlYlcnFYBo1MQ0LD9OXgUTH9yCKEC6iks5r0k5jgaJpZM4NLy9J .

-- Jun Xiong, Ph.D. U.S. Geological Survey 2255 North Gemini Drive, Flagstaff, AZ, USA. Office: 928-556-7215

dorukozturk commented 7 years ago

For big data purposes we did a very experimental Geotrellis integration. We ingested ~2TB of data which was a global 30m NDVI product which worked out pretty well. It had it's pros and cons. Ingesting was little challening but at the same time we had all the control since Geotrellis is an open source project.

suredream commented 7 years ago

I see. Good to know. I will go to check Geotrellis a bit to see the performance. Later.

aashish24 commented 7 years ago

@suredream any update on this issue (Should we keep it open)

aashish24 commented 7 years ago

@suredream I would like to mention that it would be worth considering google earth engine but I would like to discuss it with our team first.

dementiev commented 6 years ago

@dorukozturk do you have any plans to opensource geotrellis example as well?

dorukozturk commented 6 years ago

@dementiev You should definitely take a look at https://github.com/locationtech-labs/geopyspark They have geonotebook integration with a python interface to geotrellis.