Welcome to eemont!#
GitHub:davemlz/eemont
Documentation:https://eemont.readthedocs.io/
Conda-Forge:https://anaconda.org/conda-forge/eemont
Tutorials:davemlz/eemont
Table of Contents
Overview#
Google Earth Engine is a cloud-based service for geospatial processing of vector and raster data. The Earth Engine platform has aJavaScript and a Python API with different methods to process geospatial objects. Google Earth Engine also provides aHUGE PETABYTE-SCALE CATALOG of raster and vector data that users can process online (e.g. Landsat Missions Image Collections, Sentinel Missions Image Collections, MODIS Products Image Collections, World Database of Protected Areas, etc.). The eemont package extends theGoogle Earth Engine Python API with pre-processing and processing tools for the most used satellite platforms by adding utility methods for differentEarth Engine Objects that are friendly with the Python method chaining.
Google Earth Engine Community: Developer Resources#
The eemont Python package can be found in theEarth Engine Community: Developer Resources together with other awesome resources such asgeemap andrgee.
Additional Resources#
If you like eemont, you might be interested in…
Awesome Spectral Indices for GEE: A ready-to-use curated list of spectral indices for Google Earth Engine.
spectral: Awesome Spectral Indices for the Google Earth Engine JavaScript API (Code Editor).
eeExtra: A ninja Python package behind rgee, rgeeExtra and eemont.
rgeeExtra: High-level functions to process spatial and simple Earth Engine objects.
How does it work?#
The eemont python package extends the following Earth Engine classes:
New utility methods and constructors are added to above-mentioned classes in order to create a more fluid code by being friendly with the Python method chaining. These methods are mandatory for some pre-processing and processing tasks (e.g. clouds masking, shadows masking, image scaling, spectral indices computation, etc.), and they are presented as simple functions that give researchers, students and analysts the chance to analyze data with far fewer lines of code.
Look at this simple example where aSentinel-2 Surface Reflectance Image Collection is pre-processed and processed in just one step:
importee,eemontee.Authenticate()ee.Initialize()point=ee.Geometry.PointFromQuery('Cali, Colombia',user_agent='eemont-example')# Extended constructorS2=(ee.ImageCollection('COPERNICUS/S2_SR').filterBounds(point).closest('2020-10-15')# Extended (pre-processing).maskClouds(prob=70)# Extended (pre-processing).scaleAndOffset()# Extended (pre-processing).spectralIndices(['NDVI','NDWI','BAIS2']))# Extended (processing)
And just like that, the collection was pre-processed, processed and ready to be analyzed!
Installation#
Install the latest eemont version from PyPI by running:
pipinstalleemont
Upgrade eemont by running:
pipinstall-Ueemont
Install the development version from GitHub by running:
pipinstallgit+https://github.com/davemlz/eemont
Install the latest eemont version from conda-forge by running:
condainstall-cconda-forgeeemont
Features#
Let’s see some of the main features of eemont and how simple they are compared to the GEE Python API original methods:
Overloaded Operators#
The following operators are overloaded: +, -, *, /, //, %, **, <<, >>, &, |, <, <=, ==, !=, >, >=, -, ~. (and you can avoid theee.Image.expression()
method!)
GEE Python API | eemont-style |
---|---|
ds='COPERNICUS/S2_SR'S2=(ee.ImageCollection(ds).first())defscaleImage(img):scaling=img.select('B.*')x=scaling.multiply(0.0001)scaling=img.select(['AOT','WVP'])scaling=scaling.multiply(0.001)x=x.addBands(scaling)notScaling=img.select(['SCL','TCI.*','MSK.*','QA.*']))returnx.addBands(notScaling)S2=scaleImage(S2)exp='2.5*(N-R)/(N+(6*R)-(7.5*B)+1)'imgDict={'N':S2.select('B8'),'R':S2.select('B4'),'B':S2.select('B2')}EVI=S2.expression(exp,imgDict) | ds='COPERNICUS/S2_SR'S2=(ee.ImageCollection(ds).first().scale())N=S2.select('B8')R=S2.select('B4')B=S2.select('B2')EVI=2.5*(N-R)/(N+(6*R)-(7.5*B)+1) |
Clouds and Shadows Masking#
Masking clouds and shadows can be done using eemont with just one method:maskClouds()
!
GEE Python API | eemont-style |
---|---|
1ds='LANDSAT/LC08/C01/T1_SR' 2 3defmaskCloudsShadows(img): 4c=(1<<3) 5s=(1<<5) 6qa='pixel_qa' 7qa=img.select(qa) 8cm=qa.bitwiseAnd(c).eq(0) 9sm=qa.bitwiseAnd(s).eq(0)10mask=cm.And(sm)11returnimg.updateMask(mask)1213(ee.ImageCollection(ds)14.map(maskCloudsShadows)) | 1ds='LANDSAT/LC08/C01/T1_SR'23(ee.ImageCollection(ds)4.maskClouds()) |
Image Scaling and Offsetting#
Scaling and offsetting can also be done using eemont with just one method:scale()
!
GEE Python API | eemont-style |
---|---|
1defscaleBands(img): 2scaling=img.select([ 3'NDVI', 4'EVI', 5'sur.*' 6]) 7x=scaling.multiply(0.0001) 8scaling=img.select('.*th') 9scaling=scaling.multiply(0.01)10x=x.addBands(scaling)11notScaling=img.select([12'DetailedQA',13'DayOfYear',14'SummaryQA'15])16returnx.addBands(notScaling)1718ds='MODIS/006/MOD13Q1'1920(ee.ImageCollection(ds)21.map(scaleBands)) | 1ds='MODIS/006/MOD13Q1'23(ee.ImageCollection(ds)4.scale()) |
Complete Preprocessing#
The complete preprocessing workflow (Masking clouds and shadows, and image scaling and offsetting) can be done using eemont with just one method:preprocess()
!
GEE Python API | eemont-style |
---|---|
1ds='LANDSAT/LC08/C01/T1_SR' 2 3defmaskCloudsShadows(img): 4c=(1<<3) 5s=(1<<5) 6qa='pixel_qa' 7qa=img.select(qa) 8cm=qa.bitwiseAnd(c).eq(0) 9sm=qa.bitwiseAnd(s).eq(0)10mask=cm.And(sm)11returnimg.updateMask(mask)1213defscaleBands(img):14scaling=img.select('B[1-7]')15x=scaling.multiply(0.0001)16scaling=img.select([17'B10','B11'18])19scaling=scaling.multiply(0.1)20x=x.addBands(scaling)21notScaling=img.select([22'sr_aerosol',23'pixel_qa',24'radsat_qa'25])26returnx.addBands(notScaling)2728(ee.ImageCollection(ds)29.map(maskCloudsShadows)30.map(scaleBands)) | 1ds='LANDSAT/LC08/C01/T1_SR'23(ee.ImageCollection(ds)4.preprocess()) |
Spectral Indices#
Do you need to compute several spectral indices? Use theindex()
method! A lot of built-in vegetation, burn, water, snow, drought and kernel indices can be computed:
GEE Python API | eemont-style |
---|---|
ds='LANDSAT/LC08/C01/T1_SR'defscaleImage(img):scaling=img.select('B[1-7]')x=scaling.multiply(0.0001)scaling=img.select(['B10','B11'])scaling=scaling.multiply(0.1)x=x.addBands(scaling)notScaling=img.select(['sr_aerosol','pixel_qa','radsat_qa']))returnx.addBands(notScaling)defaddIndices(img):x=['B5','B4']a=img.normalizedDifference(x)a=a.rename('NDVI')x=['B5','B3']b=img.normalizedDifference(x)b=b.rename('GNDVI')x=['B3','B6']c=img.normalizedDifference(x)c=b.rename('NDSI')returnimg.addBands([a,b,c])(ee.ImageCollection(ds).map(scaleImage).map(addIndices)) | ds='LANDSAT/LC08/C01/T1_SR'(ee.ImageCollection(ds).scale().index(['NDVI','GNDVI','NDSI'])) |
The list of available indices can be retrieved by running:
eemont.listIndices()
Information about the indices can also be checked:
indices=eemont.indices()indices.BAIS2.formulaindices.BAIS2.reference
Closest Image to a Specific Date#
Struggling to get the closest image to a specific date? Here is the solution: theclosest()
method!
GEE Python API | eemont-style |
---|---|
1ds='COPERNICUS/S5P/OFFL/L3_NO2' 2 3xy=[-76.21,3.45] 4poi=ee.Geometry.Point(xy) 5 6date=ee.Date('2020-10-15') 7date=date.millis() 8 9defsetTimeDelta(img):10prop='system:time_start'11prop=img.get(prop)12prop=ee.Number(prop)13delta=prop.subtract(date)14delta=delta.abs()15returnimg.set(16'dateDist',17delta)1819(ee.ImageCollection(ds)20.filterBounds(poi)21.map(setTimeDelta)22.sort('dateDist')23.first()) | 1ds='COPERNICUS/S5P/OFFL/L3_NO2'23xy=[-76.21,3.45]4poi=ee.Geometry.Point(xy)56(ee.ImageCollection(ds)7.filterBounds(poi)8.closest('2020-10-15')) |
Time Series By Regions#
The JavaScript API has a method for time series extraction (included in the ui.Chart module), but this method is missing in the Python API… so, here it is!
PD: Actually, there are two methods that you can use:getTimeSeriesByRegion()
andgetTimeSeriesByRegions()
!
f1=ee.Feature(ee.Geometry.Point([3.984770,48.767221]).buffer(50),{'ID':'A'})f2=ee.Feature(ee.Geometry.Point([4.101367,48.748076]).buffer(50),{'ID':'B'})fc=ee.FeatureCollection([f1,f2])S2=(ee.ImageCollection('COPERNICUS/S2_SR').filterBounds(fc).filterDate('2020-01-01','2021-01-01').maskClouds().scale().index(['EVI','NDVI']))# By Regionts=S2.getTimeSeriesByRegion(reducer=[ee.Reducer.mean(),ee.Reducer.median()],geometry=fc,bands=['EVI','NDVI'],scale=10)# By Regionsts=S2.getTimeSeriesByRegions(reducer=[ee.Reducer.mean(),ee.Reducer.median()],collection=fc,bands=['EVI','NDVI'],scale=10)
Constructors by Queries#
Don’t you have the coordinates of a place? You can construct them by using queries!
usr='my-eemont-query-example'seattle_bbox=ee.Geometry.BBoxFromQuery('Seattle',user_agent=usr)cali_coords=ee.Feature.PointFromQuery('Cali, Colombia',user_agent=usr)amazonas_river=ee.FeatureCollection.MultiPointFromQuery('Río Amazonas',user_agent=usr)
Supported Platforms#
The Supported Platforms for each method can be found in the eemont documentation.
Masking clouds and shadows supports Sentinel Missions (Sentinel-2 SR and Sentinel-3), Landsat Missions (SR products) and some MODIS Products. Check all details in User Guide > Masking Clouds and Shadows > Supported Platforms.
Image scaling supports Sentinel Missions (Sentinel-2 and Sentinel-3), Landsat Missions and most MODIS Products. Check all details in User Guide > Image Scaling > Supported Platforms.
Spectral indices computation supports Sentinel-2 and Landsat Missions. Check all details in User Guide > Spectral Indices > Supported Platforms.
Getting the closest image to a specific date and time series supports all image collections with the
system:time_start
property.
License#
The project is licensed under the MIT license.
Contributing#
Contributions to eemont are welcome! Here you will find how to do it:
Bugs: If you find a bug, please report it by opening an issue. if possible, please attach the error, code, version, and other details.
Fixing Issues: If you want to contributte by fixing an issue, please check the eemont issues: contributions are welcome for open issues with labels
bug
andhelpwanted
.Enhancement: New features and modules are welcome! You can check the eemont issues: contributions are welcome for open issues with labels
enhancement
andhelpwanted
.Documentation: You can add examples, notes and references to the eemont documentation by using the NumPy Docstrings of the eemont documentation, or by creating blogs, tutorials or papers.
Contribution Steps#
First, fork theeemont repository and clone it to your local machine. Then, create a development branch:
gitcheckout-bname-of-dev-branch
eemont is divided according to Earth Engine classes, and you will find a module for each class (e.g.imagecollection.py
). Look for the required class as follows:
ee.Feature:
feature.py
ee.FeatureCollection:
featurecollection.py
ee.Geometry:
geometry.py
ee.Image:
image.py
ee.ImageCollection:
imagecollection.py
Thecommon.py
is used for methods that can be used for more than one Earth Engine class.
When creating new features, please start with theself
argument and add the corresponding decorator (@extend()
from theextending
module). Check this example:
from.extendingimportextend@extend(ee.image.Image,static=False)defmy_new_method(self,other):'''Returns the addition of and image and a float. Parameters ---------- self : ee.Image [this] Image to add. other : float Float to add. Returns ------- ee.Image Addition of an ee.Image and a float. Examples -------- >>> import ee, eemont >>> ee.Initialize() >>> img = ee.Image(0).my_new_method(other = 3.14) '''returnself.add(other)
By using the@extend()
decorator, themy_new_method()
method is added to theee.Image
class. If you want to add a static method, please set thestatic
argument toFalse
. Look for the required class as follows:
ee.Feature:
ee.feature.Feature
ee.FeatureCollection:
ee.featurecollection.FeatureCollection
ee.Geometry:
ee.geometry.Geometry
ee.Image:
ee.image.Image
ee.ImageCollection:
ee.imagecollection.ImageCollection
ee.List:
ee.ee_list.List
ee.Number:
ee.ee_number.Number
Remember to useBlack!
In order to test additions, you can usepytest
over thetests
folder:
pytesttests
This will automatically test all modules for the available satellite platforms through eemont. If you have added a new feature, please include it in the tests.
To test across different Python versions, please usetox
.
Now it’s time to commit your changes and push your development branch:
gitadd.gitcommit-m"Description of your work"gitpushoriginname-of-dev-branch
And finally, submit a pull request.
How to cite#
Do you like using eemont and think it is useful? Share the love by citing it!:
Montero,D.,(2021).eemont:APythonpackagethatextendsGoogleEarthEngine.JournalofOpenSourceSoftware,6(62),3168,https://doi.org/10.21105/joss.03168
If required, here is the BibTex!:
@article{Montero2021,doi={10.21105/joss.03168},url={https://doi.org/10.21105/joss.03168},year={2021},publisher={TheOpenJournal},volume={6},number={62},pages={3168},author={DavidMontero},title={eemont:APythonpackagethatextendsGoogleEarthEngine},journal={JournalofOpenSourceSoftware}}
Artists#
David Montero Loaiza: Lead Developer of eemont and eeExtra.
César Aybar: Lead Developer of rgee and eeExtra.
Aaron Zuspan: Plus Codes Constructors and Methods, Panchromatic Sharpening and Histogram Matching Developer.
Credits#
Special thanks toJustin Braaten for featuring eemont in tutorials and the GEE Community: Developer Resources Page, toCésar Aybar for the formidable help with Awesome Spectral Indices for GEE and to the JOSS Review Team (Katy Barnhart,Jayaram Hariharan,Qiusheng Wu andPatrick Gray) for the comments, suggestions and contributions!