Comments (6)
davemlz
commented on June 3, 2024
1
Wow, @aazuspan! geeSharp is amazing! Please, do it! :)
Here are some ideas:
- This method can be implemented for ee.Image and ee.ImageCollection classes. Here is the template for private functions that I use when a method works for both classes:
def _panSharpen(self, method, ...):
"""Computes pan sharpening...
Parameters
----------
self : ee.Image | ee.ImageCollection
Image or image collection to pan sharpen.
method : string
Method used to compute pan sharpening.\n
Available options:
- 'GS' : Use Gram-Schmidt.
- 'PCA' : Use Principal Component Analysis.
- ...
Returns
-------
ee.Image | ee.ImageCollection
Pan sharpened image or image collection.
"""
def L7(img):
panSharpened = "something..."
return panSharpened
def L8(img):
panSharpened = "something..."
return panSharpened
lookup = {
"LANDSAT/LC08/C01/T1_TOA": L8,
"LANDSAT/LC08/C01/T1_RT_TOA": L8,
"LANDSAT/LC08/C01/T2_TOA": L8,
"LANDSAT/LE07/C01/T1_TOA": L7,
"LANDSAT/LE07/C01/T1_RT_TOA": L7,
"LANDSAT/LE07/C01/T2_TOA": L7
}
platformDict = _get_platform_STAC(self)
if platformDict["platform"] not in list(lookup.keys()):
warnings.warn("This platform is not supported for pan sharpening.")
return self
else:
if isinstance(self, ee.image.Image):
panSharpened = lookup[platformDict["platform"]](self)
elif isinstance(self, ee.imagecollection.ImageCollection):
panSharpened = self.map(lookup[platformDict["platform"]])
return panSharpened This template is created in the common.py module and then imported into image.py and imagecollection.py modules (here an example for ee.ImageCollection):
from .common import _panSharpen
@extend(ee.imagecollection.ImageCollection)
def panSharpen(self, method = "GS", ...):
"""Computes pan sharpening...
Parameters
----------
self : ee.ImageCollection
Image Collection to pan sharpen.
method : string
Method used to compute pan sharpening.\n
Available options:
- 'GS' : Use Gram-Schmidt.
- 'PCA' : Use Principal Component Analysis.
- ...
Returns
-------
ee.ImageCollection
Pan sharpened image collection.
Examples
----------
>>> ...
"""
return _panSharpen(self, method, ...)- I see that you also have a whole module for computing QA, that's awesome! Here is an idea to compute the QA while computing the
panSharpen():
- Add a new argument to the method (
qa). - Now, since the QA is an ee.Number (or an ee.Array), the best way to add it to the image would be setting it as a property of the resulting image:
def _panSharpen(self, method, qa...):
"""Computes pan sharpening...
Parameters
----------
self : ee.ImageCollection
Image Collection to pan sharpen.
method : string
Method used to compute pan sharpening.\n
Available options:
- 'GS' : Use Gram-Schmidt.
- 'PCA' : Use Principal Component Analysis.
- ...
qa: string
Method used to compute QA.\n
Available options:
- 'MSE'
- 'PSNR'
- ...
Returns
-------
ee.ImageCollection
Pan sharpened image collection.
"""
def L7(img):
panSharpened = "something..."
QA = "something..."
return panSharpened.set({"eemont:PANSHARPEN_QA": QA})
def L8(img):
panSharpened = "something..."
QA = "something..."
return panSharpened.set({"eemont:PANSHARPEN_QA": QA})
lookup = {
"LANDSAT/LC08/C01/T1_TOA": L8,
"LANDSAT/LC08/C01/T1_RT_TOA": L8,
"LANDSAT/LC08/C01/T2_TOA": L8,
"LANDSAT/LE07/C01/T1_TOA": L7,
"LANDSAT/LE07/C01/T1_RT_TOA": L7,
"LANDSAT/LE07/C01/T2_TOA": L7
}
platformDict = _get_platform_STAC(self)
if platformDict["platform"] not in list(lookup.keys()):
warnings.warn("This platform is not supported for pan sharpening.")
return self
else:
if isinstance(self, ee.image.Image):
panSharpened = lookup[platformDict["platform"]](self)
elif isinstance(self, ee.imagecollection.ImageCollection):
panSharpened = self.map(lookup[platformDict["platform"]])
return panSharpened - User Guide: Since this method is actually a whole set of methods, I think a page in the User Guide explaining it would be amazing! (just if you agree).
If you agree, please go to ./docs/guide/ and in there please create a new file (panSharpen.rst) with the user guide for this method :) feel free to use any of the other guides as an example and don't forget to add at the beginning your name: User guide created by Aaaron Zuspan :)
- Tutorials: If you want, create as many tutorials as you wish! 🚀
That's all! Feel free to do it :)
Let me know if you need anything else or if you need help with the template!
Cheers!
from eemont.
aazuspan
commented on June 3, 2024
Thanks for putting those templates together! I wasn't sure the best way to identify the platform and bands, but that looks perfect. And adding the quality values as image properties is genius!
I'll get started on adding these features and creating a guide and tutorials :)
from eemont.
davemlz
commented on June 3, 2024
That's perfect! I'm going to assign you this Issue :)
Let me know if you need anything! 🚀
Cheers!
from eemont.
aazuspan
commented on June 3, 2024
I've got a question for you, @davemlz! Some of the pan-sharpening algorithms need to match histograms between images. I was going to just write a private function to handle that in common.py, but I thought maybe it would be worth adding to eemont as a separate method: ee.Image.matchHistogram(other). What do you think?
Noel Gorelick wrote an awesome tutorial for histogram matching in JS that could be adapted pretty quickly. The only tricky thing is that each band in the source image has to be matched to a corresponding band in the target image. No problem if they're from the same collection, but matching a Landsat-8 to a Sentinel-2 or a NAIP image, for example, would be harder.
If you think that's worth adding, would you rather I make a separate issue/PR or include it with this one?
from eemont.
davemlz
commented on June 3, 2024
Hi, @aazuspan!
Yes, that one is a nice feature and would be amazing to have it in eemont! 🚀 I think that an ee.Image.matchHistogram(other) would be very nice! But I also think this one should be more "flexible": Letting people decide which bands to get from each image, let them decide how many bands, also which images to use, and so on. Something like ee.Image.matchHistogram(targetImage, sourceBands, targetBands).
Let me know if that is possible! :) And yes, I think it could be better to record it in another Issue, but you can submit everything in one single PR if you want!
Cheers!
from eemont.
aazuspan
commented on June 3, 2024
Cool! I think being able to specify bands for histogram matching is a good solution, and it shouldn't be too hard to implement. I'll open a new issue for that.
from eemont.
Related Issues (20)
- [Suggestion] Pin requirement versions (specifically python-box)
- Update `tasseledCap` docstrings HOT 1
- Extended methods for the ee.Image class cannot be thrown to map HOT 4
- Would it be nice to expose the s2cloudless algorithm to Sentinel-2 Level 1C data? HOT 1
- ee.Geometry.Point does not accept varargs after importing eemont HOT 4
- COPERNICUS/S2_HARMONIZED HOT 3
- MODIS/061/MOD09A1 HOT 4
- What visParams should i use for al lthe indices available in spectral indices? HOT 1
- index() function not working HOT 1
- getTimeSeriesByRegion returns empty and repeated dates in dataframe HOT 1
- MODIS/061/MOD09A1 HOT 2
- tasseled cap transformation for COPERNICUS/S2_SR HOT 1
- Pixel calculation after cloud masking HOT 3
- Support for computing indices for Sentinel-1 HOT 3
- "system:id" disappears following maskClouds() that subsequently leads to the error of scaleAndOffset() HOT 1
- some methods and modules do not work HOT 5
- getTimeSeriesByRegion returns repeated and empty dates in time series HOT 2
- preprocess() left quite a few cloud shaddow HOT 1
- Timeseries in Sentinel S1 Index HOT 11
- .preprocess() failed for the new COPERNICUS/S2_SR_HARMONIZED HOT 4
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from eemont.