- Notifications
You must be signed in to change notification settings - Fork538
Rasterio reads and writes geospatial raster datasets
License
rasterio/rasterio
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
Rasterio reads and writes geospatial raster data.
Geographic information systems use GeoTIFF and other formats to organize andstore gridded, or raster, datasets. Rasterio reads and writes these formats andprovides a Python API based on N-D arrays.
Rasterio 1.4 works with Python >= 3.9, Numpy >= 1.24, and GDAL >= 3.5. Officialbinary packages for Linux, macOS, and Windows with most built-in format driversplus HDF5, netCDF, and OpenJPEG2000 are available on PyPI.
Read the documentation for more details:https://rasterio.readthedocs.io/.
Here's an example of some basic features that Rasterio provides. Three bandsare read from an image and averaged to produce something like a panchromaticband. This new band is then written to a new single band TIFF.
importnumpyasnpimportrasterio# Read raster bands directly to Numpy arrays.#withrasterio.open('tests/data/RGB.byte.tif')assrc:r,g,b=src.read()# Combine arrays in place. Expecting that the sum will# temporarily exceed the 8-bit integer range, initialize it as# a 64-bit float (the numpy default) array. Adding other# arrays to it in-place converts those arrays "up" and# preserves the type of the total array.total=np.zeros(r.shape)forbandinr,g,b:total+=bandtotal/=3# Write the product as a raster band to a new 8-bit file. For# the new file's profile, we start with the meta attributes of# the source file, but then change the band count to 1, set the# dtype to uint8, and specify LZW compression.profile=src.profileprofile.update(dtype=rasterio.uint8,count=1,compress='lzw')withrasterio.open('example-total.tif','w',**profile)asdst:dst.write(total.astype(rasterio.uint8),1)
The output:
Rasterio gives access to properties of a geospatial raster file.
withrasterio.open('tests/data/RGB.byte.tif')assrc:print(src.width,src.height)print(src.crs)print(src.transform)print(src.count)print(src.indexes)# Printed:# (791, 718)# {u'units': u'm', u'no_defs': True, u'ellps': u'WGS84', u'proj': u'utm', u'zone': 18}# Affine(300.0379266750948, 0.0, 101985.0,# 0.0, -300.041782729805, 2826915.0)# 3# [1, 2, 3]
A rasterio dataset also provides methods for getting read/write windows (likeextended array slices) given georeferenced coordinates.
withrasterio.open('tests/data/RGB.byte.tif')assrc:window=src.window(*src.bounds)print(window)print(src.read(window=window).shape)# Printed:# Window(col_off=0.0, row_off=0.0, width=791.0000000000002, height=718.0)# (3, 718, 791)
Rasterio's command line interface, named "rio", is documented atcli.rst. Itsrioinsp command opens the hood of any raster dataset so you can poke aroundusing Python.
$ rio insp tests/data/RGB.byte.tifRasterio 0.10 Interactive Inspector (Python 3.4.1)Type "src.meta", "src.read(1)", or "help(src)" for more information.>>> src.name'tests/data/RGB.byte.tif'>>> src.closedFalse>>> src.shape(718, 791)>>> src.crs{'init': 'epsg:32618'}>>> b, g, r= src.read()>>> bmasked_array(data = [[-- -- -- ..., -- -- --] [-- -- -- ..., -- -- --] [-- -- -- ..., -- -- --] ..., [-- -- -- ..., -- -- --] [-- -- -- ..., -- -- --] [-- -- -- ..., -- -- --]], mask = [[ True True True ..., True True True] [ True True True ..., True True True] [ True True True ..., True True True] ..., [ True True True ..., True True True] [ True True True ..., True True True] [ True True True ..., True True True]], fill_value = 0)>>> np.nanmin(b), np.nanmax(b), np.nanmean(b)(0, 255, 29.94772668847656)Rio provides the ability to create subcommands using plugins. Seecli.rstfor more information on building plugins.
See theplugin registryfor a list of available plugins.
The primary forum for questions about installation and usage of Rasterio ishttps://rasterio.groups.io/g/main. The authors and other users will answerquestions when they have expertise to share and time to explain. Please takethe time to craft a clear question and be patient about responses.
Please do not bring these questions to Rasterio's issue tracker, which we wantto reserve for bug reports and other actionable issues.
SeeCONTRIBUTING.rst.
Seedocs/.
SeeLICENSE.txt.
The rasterio project was begun at Mapbox and was transferred to the rasterio Github organization in October 2021.
SeeAUTHORS.txt.
SeeCHANGES.txt.
Seehere.
About
Rasterio reads and writes geospatial raster datasets