Reclassify raster in QGIS with GDAL raster calc

Question

I'm trying to reclassify some categorical raster data in QGIS using GDAL raster calculator. My input Geotiff has four values, [26, 27, 31, 32] which I want reclassified to [1.25, 2.5, 3.75, 5] respectively.

I want to use GDAL raster calc so I can use GDAL's lossless compression creation options(the input raster is quite large).

The logic I've tried is the following, where inputraster.tif = A

(A=26)*1.25 + (A=27)*2.5 + (A=31)*3.75 + (A=32)*5

Where am I going wrong? I'm unfamiliar with Numpy but from the examples I've found in the docs and online this logic seems right, but the process crashes immediately.

Answer 1

I tried it out following command:

gdal_calc.py -A raster_to_classify2.tif --calc "(A==26) * 1.25 + (A==27) * 2.5 + (A==31) * 3.75 + (A==32) * 5." --type Float32 --outfile raster_classified.tif

with my raster_to_classify2.tif (aleatory values 26, 27, 31, 32) and it worked.

Answer 2

You could also use gdal_reclassify as described here

python gdal_reclassify.py source_dataset.tif destination_dataset.tif -c "==26, ==27, ==31, ==32" -r "1.25, 2.5, 3.75, 5" -n true