the one script#
Below is a sample “one big script” to process a set of KH-9 mapping camera images from start (as .tgz files) to
finish (georeferenced, masked DEM and Orthomosaic).
All you have to do is copy the script below, save it to a file, and fill in these lines:
local_crs = None # need to fill in the crs to use
glacmask = None # fill in path to optional glacier mask
landmask = None # fill in path to optional land mask
out_name = '' # fill in prefix for final files
Then, run the script:
python the_one_script.py
the script#
#!/usr/bin/env python
from glob import glob
from spymicmac import data, preprocessing, micmac, register
local_crs = None # need to fill in the crs to use
glacmask = None # fill in path to optional glacier mask
landmask = None # fill in path to optional land mask
out_name = '' # fill in prefix for final files
imlist = [fn.split('.tgz')[0] for fn in glob('*.tgz')]
# download + reproject cop30 DEM
data.download_cop30_vrt(imlist, crs=local_crs)
# preprocess images
preprocessing.preprocess_kh9_mc(
skip='balance',
nproc='max',
blend=True,
add_sfs=True,
res_low=400,
res_high=8000,
add_params=True
)
# create the relative dem/orthophoto
micmac.malt(
'OIS.*tif',
'Relative',
dirmec='MEC-Relative',
zoomf=2,
cost_trans=4,
szw=3,
regul=0.1
)
# register the images to the reference DEM
register.register_relative(
'MEC-Relative',
'Copernicus_DSM_ell.tif',
glacmask=glacmask,
landmask=landmask,
density=250,
strategy='chebyshev'
)
# create the absolute dem/orthophotos
micmac.malt(
'OIS.*tif',
'TerrainFinal',
dirmec='MEC-Malt',
zoomf=1,
cost_trans=4,
szw=3,
regul=0.1,
resol_terr=10,
resol_ort=2
)
# create the orthomosaic
micmac.tawny('MEC-Malt')
# clean up the outputs
micmac.post_process(
projstr=local_crs,
out_name=out_name,
dirmec='MEC-Malt',
do_ortho=True,
ind_ortho=True
)