Preserving coordinates when converting TIFF image to numpy ndarray

Question

I have satellite image that I want to convert into numpy array (and then to Pandas). I already know how to do that, but the problem is that it does not preserve the coordinate data.

This is how I do it now:

import xarray
src=rasterio.open('14072020.tif')
array = src.read()
pd.DataFrame(array.reshape([13,-1]).T)

I have also tried to follow this answer (keeping the coordinate system of raster files in the resulting raster file after operation with numpy) but I always get this error:

NameError: name 'gdal_array' is not defined

which does not allow me to open the image.

My end goal is to have Pandas table that contains the bands values together with the coordinate values.

• snowman2: Does this answer your question? Convert raster to CSV with lat, lon, and value columns

Answer 1

You can use the open_rasterio and to_dataframe methods to accomplish that.

import rioxarray
rds = rioxarray.open_rasterio("file.tif")
rds.to_dataframe()

See also: https://gis.stackexchange.com/a/358057/144357

Answer 2

I would read with gdal and dump that to a numpy array. ReadAsArray().

Coordinates and projection are obtained using the GetGeotransform() SetGeotransform() and the GetProjection() and SetProjection() applied to the data set.

Example here. https://pcjericks.github.io/py-gdalogr-cookbook/raster_layers.html