gdal_translate with custom corner coordinates in python

Question

I wish to convert this Python code using a gdal command line tool to using the gdal python library. I am using it to convert some PNG satellite images I am downloading to georeferenced TIFFs. I manually specify the bounding box of the images when I download them (in EPSG:3857), so I know all the info about the images, I just wish to store all the info in a single georeferenced TIFF file.

My current solution is to use gdal_translate called by subprocess but apparently this can all be done natively in python, so I would prefer to do that rather than rely on this hacked-together workaround.

from osgeo import gdal
import subprocess
import shlex    

# ll_x etc... are all corner coords I calculate before hand 
subprocess.run(shlex.split("gdal_translate -of GTiff -a_ullr {} {} {} {} -a_srs {} {} {}".format(
    ll_x, ur_y, ur_x, ll_y, 
    "EPSG:3857", 
    "input_sat_image.png", 
    "referenced_sat.tif")))

I have seen individual questions that cover pieces of this, but I am struggling to put them all together.

Answer 1

I might be wrong, but maybe you can try this way :

from osgeo import gdal 
import osr

inputImage = YOURNAME 
outputImage = YOURNAME

dataset = gdal.Open(inputImage) 
I = dataset.ReadAsArray(0,0,dataset.RasterXSize,dataset.RasterYSize)

outdataset = gdal.GetDriverByName('GTiff') 
output_SRS = osr.SpatialReference() 
output_SRS.ImportFromEPSG(3857) 
outdataset = outdataset.Create(outputImage,dataset.RasterXSize,dataset.RasterYSize,I.shape[0],gdal_Float32) 
for nb_band in range(I.shape[0]):
    outdataset.GetRasterBand(nb_band+1).WriteArray(I[nb_band,:,:])

gcp_list = [] 
gcp_list.append(gdal.GCP(lon,lat,alt,col,row)) #Repeat this step for as many GCP as you want (there is a limit but I don't know the number of GCPs)

outdataset.SetProjection(srs.ExportToWkt()) 
wkt = outdataset.GetProjection() 
outdataset.SetGCPs(gcp_list,wkt)

outdataset = None

It worked for me, but it was in a specific case (and I did not need a very precise result...). I think you might have a problem with the sizes, but I actually don't know how to solve it.