An error occurred while loading the file. Please try again.
-
Pierre-Antoine Rouby authored2e158f4c
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
#import argparse
import glob, os, sys, warnings
import subprocess
from itertools import groupby
import gdal, ogr, osr
import otbApplication as otb
from mtdUtils import getBufferedGeographicExtent, getRasterEPSG, getGeoJSONExtent_WGS84, getRasterExtentAsGeometry
import xml.etree.ElementTree as ET
import multiprocessing as mp
from path import Path
import tempfile
import zipfile
from shutil import unpack_archive
import elevation
# TODO:
# - rasterize vector on
# - check possibilities to update filter outcore to make filtering on a temporal moving window
def download_srtm1(roi_vector, outfile, cache_dir=Path('~/.moringa/cache').expanduser(), max_download_tiles=9,
overwrite=False):
"""
Download SRTM1 (i.e. at ~30m) from AWS
Parameters
----------
roi_vector
outfile
cache_dir
max_download_tiles: int
For more than 9 it may be blocked, see https://github.com/bopen/elevation
Returns
-------
"""
# TODO:
# another possibility would be:
# 1. use elevation to specify the tiles
# 2. download from https://urs.earthdata.nasa.gov as specified by https://dwtkns.com/srtm30m/
# 3. unzip file and convert to tiff
# but an account on https://urs.earthdata.nasa.gov is needed
# see https://wiki.earthdata.nasa.gov/display/EL/How+To+Access+Data+With+Python
# More simple with curl (tested and working): https://wiki.earthdata.nasa.gov/display/EL/How+To+Access+Data+With+cURL+And+Wget
# see also https://ec.haxx.se/usingcurl/usingcurl-netrc
# tile example: http://e4ftl01.cr.usgs.gov/MEASURES/SRTMGL1.003/2000.02.11/N45E003.SRTMGL1.hgt.zip
#
# Get the tiles names :
# use https://dwtkns.com/srtm30m/srtm30m_bounding_boxes.json
# or the following code with elevation :
# datasource_root, spec = elevation.datasource.ensure_setup(cache_dir, 'SRTM1')
# bounds = elevation.datasource.build_bounds(bounds, margin='0')
# tiles_names = spec['tile_names'](*bounds)
outfile = Path(outfile)
if outfile.exists():
print('SRTM download skipped, SRTM file already exists: {}'.format(outfile))
return outfile
# bounds = gpd.read_file(roi_vector).to_crs('+init=epsg:4326').bounds.values[0, :]
# bounds = gpd.read_file(roi_vector).to_crs(4326).bounds.values[0, :]
# bounds = getSQLiteExtent_WGS84(roi_vector)
bounds = getGeoJSONExtent_WGS84(roi_vector)
#bounds = [bounds[i] for i in [1,0,3,2]]
elevation.clip(bounds, outfile, cache_dir=cache_dir, max_download_tiles=max_download_tiles)
return outfile
def download_srtm3(roi_vector, outdir, verbose=True):
# 2 solutions are available at the moment:
# https://github.com/bopen/elevation downloads from AWS tiles in a cache dir and clip to given latlon bounds
# https://www.orfeo-toolbox.org/CookBook/Applications/app_DownloadSRTMTiles.html downloads tiles from USGS in zip for given vector, raster or tile name
if verbose:
print('Downloading SRTM3...')
outdir = Path(outdir)
app = otb.Registry.CreateApplication('DownloadSRTMTiles')
app.SetParameterString('vl', roi_vector)
app.SetParameterString('tiledir', outdir)
app.ExecuteAndWriteOutput()
zipfiles = outdir.glob('*.zip')
if verbose:
print('Unzipping tiles:')
outfiles = []
with tempfile.TemporaryDirectory() as tmpdir:
tmpdir_path = Path(tmpdir)
for zf in zipfiles:
if verbose:
print('\t{}'.format(zf))
with zipfile.ZipFile(zf, "r") as j:
tmpfile = tmpdir_path / j.namelist()[0].replace('/', '')
outfile = outdir / tmpfile.name
unpack_archive(zf, tmpdir_path)
tmpfile.move(outfile)
outfiles.append(outfile)
zf.remove()
return outfiles
def download_geoid(cache_dir=Path('~/.moringa/cache').expanduser(), verbose=True):
geoid_link = 'https://gitlab.orfeo-toolbox.org/orfeotoolbox/otb-data/-/raw/master/Input/DEM/egm96.grd'
geoid_file = Path(cache_dir) / 'emg96.grd'
if not geoid_file.exists():
cmd = 'curl -o {} {}'.format(geoid_file, geoid_link)
if not verbose:
cmd + '-s -S'
subprocess.check_call(cmd, shell=True)
else:
print('Geoid file download skipped, already exists: {}'.format(geoid_file))
return geoid_file
def S1CalibOrtho(fld, epsg=None, demfld=None, geoid=None, overwrite=False):
lst = glob.glob(fld + os.sep + 'measurement' + os.sep + '*.tiff')
sarcal = []
ortho = []
out = []
for l in lst:
sarcal.append(otb.Registry.CreateApplication('SARCalibration'))
ortho.append(otb.Registry.CreateApplication('OrthoRectification'))
sarcal[-1].SetParameterString('in', l)
sarcal[-1].SetParameterString('lut', 'sigma')
sarcal[-1].Execute()
ortho[-1].SetParameterString('elev.dem', demfld)
ortho[-1].SetParameterString('elev.geoid', geoid)
ortho[-1].SetParameterString('map', 'epsg')
ortho[-1].SetParameterString('map.epsg.code', str(epsg))
ortho[-1].SetParameterInt('opt.gridspacing', 40)
ortho[-1].SetParameterInputImage('io.in', sarcal[-1].GetParameterOutputImage('out'))
out.append(l.replace('.tiff', '_calsigma_ortho.tiff'))
ortho[-1].SetParameterString('io.out', out[-1])
if (os.path.exists(out[-1]) and not overwrite):
print('Calibration and Orthorectification process skipped, file already exists:\n\t{}'.format(out[-1]))
else:
ortho[-1].ExecuteAndWriteOutput()
return out
def S1CalibOrthoWithROI(infile, outfile, roi, demfld, geoid, lut='sigma', overwrite=False, ram=1024):
if (os.path.exists(outfile) and not overwrite):
print('Calibration and Orthorectification process skipped, file already exists:\n\t{}'.format(outfile))
return outfile
else:
if checkRoiInS1Acquisition(infile, roi):
pipe = []
pipe.append(otb.Registry.CreateApplication('SARCalibration'))
pipe[-1].SetParameterString('in', infile)
pipe[-1].SetParameterString('lut', lut)
pipe[-1].SetParameterString('ram', str(ram))
pipe[-1].Execute()
pipe.append(otb.Registry.CreateApplication('OrthoRectification'))
pipe[-1].SetParameterInputImage('io.in', pipe[-2].GetParameterOutputImage('out'))
pipe[-1].SetParameterString('elev.dem', demfld)
pipe[-1].SetParameterString('elev.geoid', geoid)
pipe[-1].SetParameterString('map', 'epsg')
pipe[-1].SetParameterString('map.epsg.code', getRasterEPSG(roi))
pipe[-1].SetParameterInt('opt.gridspacing', 40)
pipe[-1].SetParameterString('opt.ram', str(ram))
pipe[-1].Execute()
pipe.append(otb.Registry.CreateApplication('Superimpose'))
pipe[-1].SetParameterInputImage('inm', pipe[-2].GetParameterOutputImage('io.out'))
pipe[-1].SetParameterString('inr', roi)
pipe[-1].SetParameterString('ram', str(ram))
pipe[-1].SetParameterString('out', outfile)
pipe[-1].ExecuteAndWriteOutput()
return outfile
else:
print('ROI not overlapping with S1 acquisition :\n\t{}'.format(infile))
return None
def checkRoiInS1Acquisition(s1file, roifile):
isIntersecting = False
dsv = gdal.Open(roifile)
roi_geom = getRasterExtentAsGeometry(dsv, to_wgs84=True)
dsi = ogr.Open(os.path.dirname(s1file) + os.sep + os.pardir + os.sep + 'preview' + os.sep + 'map-overlay.kml')
lyi = dsi.GetLayer()
for g in lyi:
in_geom = g.GetGeometryRef()
isIntersecting = in_geom.Intersects(roi_geom)
dsv = None
dsi = None
return isIntersecting
def preClip(infile, outfile, roi_buffered_vector, overwrite=False, verbose=True, ram=1024):
if Path(outfile).exists() and (not overwrite):
if verbose:
print('Pre-clip process skipped, file already exists:\n\t{}'.format(outfile))
return outfile
isIntersecting = False
dsv = ogr.Open(roi_buffered_vector)
lyv = dsv.GetLayer()
for f in lyv:
roi_geom = f.GetGeometryRef()
dsi = ogr.Open(os.path.dirname(infile) + os.sep + os.pardir + os.sep + 'preview' + os.sep + 'map-overlay.kml')
lyi = dsi.GetLayer()
for g in lyi:
in_geom = g.GetGeometryRef()
isIntersecting = in_geom.Intersects(roi_geom)
dsv = None
dsi = None
if (isIntersecting):
extroi = otb.Registry.CreateApplication('ExtractROI')
extroi.SetParameterString('in', infile)
extroi.SetParameterString('mode', 'fit')
extroi.SetParameterString('mode.fit.vect', roi_buffered_vector)
extroi.SetParameterString('out', outfile)
extroi.SetParameterString('ram', str(ram))
extroi.ExecuteAndWriteOutput()
return outfile
else:
if verbose:
print('Pre-clip process skipped, image outside ROI.')
return None
# cmd = ['obtcli_ExtractROI', '-in', infile, '-out', outfile,
# '-mode', 'fit', '-mode.fit.vect', roi_buffered_vector,
# '-ram', str(ram)]
# subprocess.check_call(cmd, shell=False)
def S1Calibration(infile, outfile, lut='sigma', overwrite=False, verbose=True, ram=1024):
if Path(outfile).exists() and (not overwrite):
if verbose:
print('Calibration process skipped, file already exists:\n\t{}'.format(outfile))
return outfile
print(lut)
sarcal = otb.Registry.CreateApplication('SARCalibration')
sarcal.SetParameterString('in', infile)
sarcal.SetParameterString('lut', lut)
sarcal.SetParameterString('out', outfile)
sarcal.SetParameterString('ram', str(ram))
sarcal.ExecuteAndWriteOutput()
return outfile
def S1Ortho(infile, outfile, epsg, demfld, geoid, roi=None, orthofit=False, overwrite=False, verbose=True, ram=1024):
if Path(outfile).exists() and (not overwrite):
if verbose:
print('Orthorectification process skipped, file already exists:\n\t{}'.format(outfile))
return outfile
if verbose:
print('DEM folder: '.format(demfld))
ortho = otb.Registry.CreateApplication('OrthoRectification')
ortho.SetParameterString('elev.dem', demfld)
ortho.SetParameterString('elev.geoid', geoid)
ortho.SetParameterString('io.in', infile)
ortho.SetParameterString('io.out', outfile)
ortho.SetParameterString('opt.ram', str(ram))
ortho.SetParameterInt('opt.gridspacing', 40)
if orthofit and (roi is not None):
print('Using mode "orthofit" for ortho-rectification')
ortho.SetParameterString('outputs.mode', 'orthofit')
ortho.SetParameterString('outputs.ortho', roi)
else:
ortho.SetParameterString('map', 'epsg')
ortho.SetParameterString('map.epsg.code', str(epsg))
ortho.ExecuteAndWriteOutput()
return outfile
def Superimpose(infile, roi, outfile, overwrite=False, verbose=True, ram=1024):
if Path(outfile).exists() and (not overwrite):
if verbose:
print('Superimpose process skipped, file already exists:\n\t{}'.format(outfile))
return outfile
app = otb.Registry.CreateApplication('Superimpose')
app.SetParameterString('inr', roi)
app.SetParameterString('inm', infile)
app.SetParameterString('out', outfile)
app.SetParameterString('ram', str(ram))
app.ExecuteAndWriteOutput()
return outfile
def MultitempFilteringOutcore(infiles, outfile, window_radius=3, overwrite=False, verbose=True, ram=1024):
if len(infiles)==0:
return None
if Path(outfile).exists() and (not overwrite):
if verbose:
print('MultitempFilteringOutcore process skipped, file already exists:\n\t{}'.format(outfile))
return outfile
app = otb.Registry.CreateApplication('MultitempFilteringOutcore')
app.SetParameterStringList('inl', infiles)
app.SetParameterString('wr', str(window_radius))
app.SetParameterString('oc', outfile)
app.SetParameterString('ram', str(ram))
app.ExecuteAndWriteOutput()
# (infiles[0].parent / 'filtered').mkdir_p()
# otb_path / 'bin' /
# cmd = ['otbcli_MultitempFilteringOutcore', '-inl'] + infiles + \
# ['-wr', str(window_radius), '-oc', outfile, '-ram', str(ram)]
# if verbose:
# print(' '.join(cmd))
# subprocess.check_call(cmd, shell=False)
return outfile
# cmd = ['otbcli_MultitempFilteringOutcore', '-inl'] + files + ['-wr', str(window_radius), '-oc',
# outcore_file]
def MultitempFilteringFilter(infiles, outdir, window_radius, outcore_file, enl_file, overwrite=False, verbose=True, ram=1024):
infiles = [Path(f) for f in infiles]
outdir = Path(outdir)
outfiles = [enl_file] + [outdir / f.name.stripext() + '_filtered.tiff' for f in infiles]
existing = [f for f in outfiles if f.exists()]
if (len(existing) > 0) and (not overwrite):
if len(existing) != len(outfiles):
raise IOError('Some outputs of MultitempFilteringFilter already exist but not all,'
'set overwrite argument to overwrite:\n\t{}'.format(existing))
if verbose:
print('MultitempFilteringFilter process skipped, files already exists:\n\t{}'.format(existing))
return outfiles
"""
app = otb.Registry.CreateApplication('MultitempFilteringFilter')
app.SetParameterStringList('inl', infiles) # list parameter not accpeted
app.SetParameterString('wr', str(window_radius))
app.SetParameterString('oc', outcore_file)
app.SetParameterString('enl', enl_file)
app.SetParameterString('filtpath', outdir)
app.SetParameterString('ram', str(10*ram))
app.ExecuteAndWriteOutput()
# infiles[0].parent.mkdir_p()
# cmd = ['otbcli_MultitempFilteringFilter', '-inl'] + infiles + ['-wr', str(window_radius), '-oc',
# outcore_file, '-enl', enl_file,
# '-filtpath', outdir, '-ram', str(ram)]
# if verbose:
# print(cmd)
# subprocess.check_call(cmd, shell=False)
"""
i = 1
for f in infiles:
pipe = []
pipe.append(otb.Registry.CreateApplication('Smoothing'))
pipe[-1].SetParameterString('in', f)
pipe[-1].SetParameterString('type', 'mean')
pipe[-1].SetParameterString('type.mean.radius', str(window_radius))
pipe[-1].SetParameterString('ram', str(ram))
pipe[-1].Execute()
pipe.append(otb.Registry.CreateApplication('BandMath'))
pipe[-1].SetParameterStringList('il', [outcore_file])
pipe[-1].AddImageToParameterInputImageList('il', pipe[-2].GetParameterOutputImage('out'))
pipe[-1].SetParameterString('exp', 'im2b1*im1b1/im1b2')
pipe[-1].SetParameterString('ram', str(ram))
pipe[-1].SetParameterString('out', outfiles[i])
pipe[-1].ExecuteAndWriteOutput()
i += 1
return outfiles
def extractROI(infiles, outdir, roi, roiname=None, ncores=12, overwrite=False, verbose=True, ram=1024):
"""
Merge same date images with average (through gdalbuildvrt) and extracts roi
Parameters
----------
infiles: list
List S1 tiff files to process
outdir: str
Output directory where are stored generated files
roi: str
Region of interest raster
roiname: str
name used as suffix of generated files: xxx_roiname.tiff
ncores: int
Number of cores used in multiprocessing
overwrite: bool
If True, overwrites existing files.
verbose: bool
Returns
-------
list
Ouput files
"""
if verbose:
print('Mosaic and ROI extraction process...')
infiles = [Path(infile) for infile in infiles]
outdir = Path(outdir)
roi = Path(roi)
# roi must be a raster over the ROI, outputs will be superimposed over this reference.
if roiname is None:
roiname = roi.name.splitext()[0]
infiles = sorted(infiles, key=lambda x: x.name[11:29])
slst = [list(i) for j, i in groupby(infiles, lambda x: x.name[11:22])]
#### build virtual rasters of same date --> layer average when loaded
vrtfiles = []
outfiles = []
proc_files = []
for l in slst:
oldd = l[0].name[14:45]
newd = l[0].name[14:29] + '-' + l[-1].name[30:45]
outfile = outdir / l[0].name.replace(oldd, newd).replace('.tiff', '_' + roiname + '.tiff')
outfiles.append(outfile)
if (not outfile.exists()) or overwrite:
vrt = outfile.replace('.tiff', '.vrt')
vrtfiles.append(vrt)
proc_files.append(outfile)
vrtcmd = ['gdalbuildvrt', vrt] + l
subprocess.call(vrtcmd, shell=False)
tiffcmd = ['gdal_translate', vrt, outfile]
subprocess.call(tiffcmd, shell=False)
return outfiles
def multitempSpeckleFilt(infiles, outdir, window_radius=3, overwrite=False, verbose=True, ram=1024):
print('Multi-temporal speckle filtering...')
infiles = [Path(f) for f in infiles]
outdir = Path(outdir)
polars = ['vh', 'vv']
sensors = ['s1a', 's1b']
outfiles = []
for polar in polars:
polar_files = [f for f in infiles if f.name.split('-')[3] == polar]
for sensor in sensors:
files = [f for f in polar_files if f.name.startswith(sensor)]
outcore_file = outdir / 'outcore_{}_{}.tiff'.format(sensor, polar)
enl_file = outdir / 'enl_{}_{}.tiff'.format(sensor, polar)
if len(files) > 0:
#### generate outcore file
if verbose:
print('\tComputing speckle filter file: {}'.format(outcore_file))
MultitempFilteringOutcore(files, outcore_file, window_radius=window_radius, overwrite=overwrite, verbose=verbose, ram=ram)
#### apply filtering: outputs are enl_file and outdir/files.name
if verbose:
print('\tFiltering images {} {} ...'.format(sensor, polar))
psoutfiles = MultitempFilteringFilter(files, outdir, window_radius,
outcore_file, enl_file , overwrite, verbose, ram=ram)
outfiles.extend([outcore_file] + psoutfiles)
return outfiles
#
# lstvh = sorted(glob.glob(fld + os.sep + '*vh*_calsigma_ortho_*.tiff'), key=lambda x: os.path.basename(x)[14:22])
# if len(lstvh) == 0:
# lstvh = sorted(glob.glob(fld + os.sep + '*.SAFE' + os.sep + 'measurement' + os.sep + '*vh*_calsigma_ortho.tiff'), key=lambda x: os.path.basename(x)[17:25])
# lstvv = sorted(glob.glob(fld + os.sep + '*vv*_calsigma_ortho_*.tiff'), key=lambda x: os.path.basename(x)[14:22])
# if len(lstvv) == 0:
# lstvv = sorted(glob.glob(fld + os.sep + '*.SAFE' + os.sep + 'measurement' + os.sep + '*vv*_calsigma_ortho.tiff'), key=lambda x: os.path.basename(x)[17:25])
#
# ofld = fld + os.sep + 'specklefilter'
# if not os.path.exists(ofld):
# os.mkdir(ofld)
# if not os.path.exists(ofld + os.sep + 'filtered'):
# os.mkdir(ofld + os.sep + 'filtered')
#
# slinksvh = [ofld + os.sep + os.path.basename(x) for x in lstvh]
# slinksvv = [ofld + os.sep + os.path.basename(x) for x in lstvv]
# [os.symlink(lstvh[i], slinksvh[i]) for i in range(len(lstvh)) if not os.path.exists(slinksvh[i])]
# [os.symlink(lstvv[i], slinksvv[i]) for i in range(len(lstvv)) if not os.path.exists(slinksvv[i])]
#
# if len(slinksvh) > 0:
# cmd = ['otbcli_MultitempFilteringOutcore', '-inl'] + slinksvh + ['-wr',str(window_radius),'-oc',fld + os.sep + 'outcore_vh.tiff']
# subprocess.call(cmd, shell=False)
# cmd = ['otbcli_MultitempFilteringFilter', '-inl'] + slinksvh + ['-wr', str(window_radius), '-oc', fld + os.sep + 'outcore_vh.tiff', '-enl', fld + os.sep + 'enl_vh.tiff']
# subprocess.call(cmd, shell=False)
# #cmd = ['otbcli_MultitempFiltering', '-inl'] + slinksvh + ['-wr', str(window_radius)]
# #subprocess.call(cmd,shell=False)
# if len(lstvv) > 0:
# cmd = ['otbcli_MultitempFilteringOutcore', '-inl'] + slinksvv + ['-wr',str(window_radius),'-oc',fld + os.sep + 'outcore_vv.tiff']
# subprocess.call(cmd, shell=False)
# cmd = ['otbcli_MultitempFilteringFilter', '-inl'] + slinksvv + ['-wr', str(window_radius), '-oc',fld + os.sep + 'outcore_vv.tiff', '-enl', fld + os.sep + 'enl_vv.tiff']
# subprocess.call(cmd, shell=False)
def separateS1AS1B(safe_files):
safe_files_s1a = []
safe_files_s1b = []
for l in safe_files:
if os.path.basename(l).startswith('S1A'):
safe_files_s1a.append(l)
elif os.path.basename(l).startswith('S1B'):
safe_files_s1b.append(l)
return safe_files_s1a,safe_files_s1b
def separateAscDesc(safe_files):
safe_files_asc = []
safe_files_des = []
for l in safe_files:
l = Path(l)
root = ET.parse(l / 'manifest.safe').getroot()
for item in root.iter():
if item.tag.endswith('pass'):
if item.text == 'ASCENDING':
safe_files_asc.append(l)
elif item.text == 'DESCENDING':
safe_files_des.append(l)
'''
if len(safe_files_asc) > 0 and not os.path.exists(fld + os.sep + 'ASC'):
os.mkdir(fld + os.sep + 'ASC')
[shutil.move(l, fld + os.sep + 'ASC/') for l in safe_files_asc]
safe_files_asc_out = glob.glob(fld + os.sep + 'ASC' + os.sep + '*.SAFE')
if len(safe_files_des) > 0 and not os.path.exists(fld + os.sep + 'DES'):
os.mkdir(fld + os.sep + 'DES')
[shutil.move(l, fld + os.sep + 'DES/') for l in safe_files_des]
safe_files_des_out = glob.glob(fld + os.sep + 'DES' + os.sep + '*.SAFE')
'''
return safe_files_asc, safe_files_des
def createS1Features(fld, base_name='S1series', out_fld=None, features=None, convert_to_db=True, by_date=False):
if out_fld is None:
out_fld = fld
lst_vh = sorted(glob.glob(fld + os.sep + '*iw-grd-vh*.tiff'), key=lambda x: os.path.basename(x)[14:22])
lst_vv = sorted(glob.glob(fld + os.sep + '*iw-grd-vv*.tiff'), key=lambda x: os.path.basename(x)[14:22])
# Check date correspondance
dates_vh = [os.path.basename(l)[14:22] for l in lst_vh]
dates_vv = [os.path.basename(l)[14:22] for l in lst_vv]
assert len(lst_vh) > 0 and len(lst_vv) > 0, "No vv/vh images found in folder."
assert dates_vh == dates_vv, "Dates for images in folder are not aligned for vh and vv polarizations."
vh_out = out_fld + os.sep + base_name + '_vh_series.tiff'
vv_out = out_fld + os.sep + base_name + '_vv_series.tiff'
with open(out_fld + os.sep + base_name + '_dates.txt', 'w') as f:
[f.write(x + '\n') for x in dates_vh]
if convert_to_db:
expr = '{' + ';'.join(['1000*log10(abs(im%db1)+1e-6)' % (i + 1) for i in range(len(lst_vh))]) + '}'
else:
expr = '{' + ';'.join(['10000*im%db1' % (i + 1) for i in range(len(lst_vh))]) + '}'
feat = []
feat.append(otb.Registry.CreateApplication('BandMathX'))
feat[-1].SetParameterStringList('il', lst_vh)
feat[-1].SetParameterString('exp', expr)
if by_date:
feat[-1].Execute()
else:
feat[-1].SetParameterString('out', vh_out)
feat[-1].SetParameterOutputImagePixelType('out', otb.ImagePixelType_int16)
feat[-1].ExecuteAndWriteOutput()
feat.append(otb.Registry.CreateApplication('BandMathX'))
feat[-1].SetParameterStringList('il', lst_vv)
feat[-1].SetParameterString('exp', expr)
if by_date:
feat[-1].Execute()
else:
feat[-1].SetParameterString('out', vv_out)
feat[-1].SetParameterOutputImagePixelType('out', otb.ImagePixelType_int16)
feat[-1].ExecuteAndWriteOutput()
if features is not None and 'ratio' in features:
expr = '{' + ';'.join(['1000 * im1b%d / (im2b%d + 1e-6)' % (i + 1, i + 1) for i in range(len(lst_vh))]) + '}'
feat.append(otb.Registry.CreateApplication('BandMathX'))
feat[-1].AddImageToParameterInputImageList('il', feat[0].GetParameterOutputImage('out'))
feat[-1].AddImageToParameterInputImageList('il', feat[1].GetParameterOutputImage('out'))
feat[-1].SetParameterString('exp', expr)
if by_date:
feat[-1].Execute()
else:
ratio_out = out_fld + os.sep + base_name + '_vh_vv_ratio_series.tiff'
feat[-1].SetParameterString('out', ratio_out)
feat[-1].SetParameterOutputImagePixelType('out', otb.ImagePixelType_int16)
feat[-1].ExecuteAndWriteOutput()
if by_date:
i = 1
cc = []
for d in dates_vh:
stack_out = out_fld + os.sep + base_name + '_' + d + '_feat.tif'
cc.append(otb.Registry.CreateApplication('BandMathX'))
[cc[-1].AddImageToParameterInputImageList('il', x.GetParameterOutputImage('out')) for x in feat]
cc[-1].SetParameterString('exp', '{' + ';'.join(['im%db%d' % (t,i) for t in range(1,len(feat)+1)]) + '}')
cc[-1].SetParameterString('out', stack_out)
cc[-1].SetParameterOutputImagePixelType('out', otb.ImagePixelType_int16)
cc[-1].ExecuteAndWriteOutput()
return
def createS1Quicklooks(fld, scale=0.1):
if not os.path.exists(fld + os.sep + 'QLK'):
os.mkdir(fld + os.sep + 'QLK')
lst = sorted(glob.glob(fld + os.sep + '*iw-grd-v*.tiff'))
for f in lst:
fout = fld + os.sep + 'QLK' + os.sep + 'qlk_' + os.path.basename(f)[11:]
rig = otb.Registry.CreateApplication('RigidTransformResample')
rig.SetParameterString('in', f)
rig.SetParameterFloat('transform.type.id.scalex', scale)
rig.SetParameterFloat('transform.type.id.scaley', scale)
rig.Execute()
dbc = otb.Registry.CreateApplication('BandMath')
dbc.AddImageToParameterInputImageList('il', rig.GetParameterOutputImage('out'))
dbc.SetParameterString('exp', '10*log10(abs(im1b1)+1e-6)')
dbc.Execute()
dcv = otb.Registry.CreateApplication('DynamicConvert')
dcv.SetParameterInputImage('in', dbc.GetParameterOutputImage('out'))
dcv.SetParameterFloat('quantile.high', 2.0)
dcv.SetParameterFloat('quantile.low', 2.0)
dcv.SetParameterInt('outmin', 1)
dcv.SetParameterInt('outmax', 255)
dcv.SetParameterString('out', fout)
dcv.SetParameterOutputImagePixelType('out', otb.ImagePixelType_uint8)
dcv.ExecuteAndWriteOutput()
return
'''
def raster_bbox(img, buf=0, outfile=None, driver='GeoJSON', outcrs=4326):
with rio.open(img) as r:
bounds = r.bounds
res = r.res
geom = box(bounds[0]-buf*res[0], bounds[1]-buf*res[1], bounds[2]+buf*res[0], bounds[3]+buf*res[1])
# df = gpd.GeoDataFrame({"id":1,"geometry":[geom]}, crs=r.crs).to_crs('+init=epsg:{}'.format(outcrs))
df = gpd.GeoDataFrame({"id": 1, "geometry": [geom]}, crs=r.crs).to_crs(outcrs)
if outfile is None:
return df
else:
df.to_file(outfile, driver=driver)
return outfile
'''
def raster_epsg(x):
d = gdal.Open(x)
proj = osr.SpatialReference(wkt=d.GetProjection())
return int(proj.GetAttrValue('AUTHORITY', 1))
def s1process(indir, outdir, epsg=None,
roi=None, buffer=100, dem=None, geoid=None, orthofit=False, lut='sigma',
direction='descending', satellite='both', step='all', force_clip=False, features=None, skip_db_conversion=False,
features_by_date=False, calib_dir=None, cache_dir=Path('~/.moringa/cache').expanduser(),
ram=1024, ncores=1, overwrite=False, verbose=True):
"""Process Sentinel 1 data
Process S1 SAFE images:
(clip to ROI extent ->) calibrate -> orthorectify -> mosaic (-> superimpose) -> filter speckle
see ortho
Parameters
----------
indir: str | list
Directory containing .SAFE directories or list of .SAFE directories.
Or the output directory of one of the processing step, e.g. 'myROI/calib'.
outdir: str
Output directory where will be stored processed data.
epsg: int
Input EPSG. If not given, EPSG is deduced from ROI.
roi: str
Region Of Interest raster file. If not specified, EPSG and DEM are expected.
buffer: int
Number of pixels extending ROI extent for SRTM cropping to avoid border effects during ortho-rectification process.
dem: str
DEM directory. Optional, if not given SRTM30m will be downloaded over ROI.
geoid str
Geoid file. If not given, emg96.grd from OTB repo is used.
orthofit: bool
If activated it will use mode orthofit with ROI in OTB OrthoRectification app.
Otherwise image is clipped before calibration step and
aligned (OTB Superimpose) on ROI grid after mosaic
direction: str
Direction of acquisition, either ascending, descending or both
step: str | list
Specify one or more processing steps. Available steps are ['all', 'clip', 'calib', 'ortho', 'mosaic', 'superimpose', 'filter'].
If orthofit=True: 'all' --> ['calib', 'ortho', 'mosaic', 'filter']
Otherwise: 'all' --> ['clip', 'calib', 'ortho', 'mosaic', 'superimpose', 'filter']
calib_dir: str
Specify calibration directory to avoid computation of calibration.
This option should not be used if the processing includes pre-clip step.
cache_dir: str
Cache directory for elevation package, it is where are stored the SRTM tiles (<50MB/tile) before crop to ROI.
ram: int
RAM limit to be used in OTB applications. Warning: ncores x ram may be used, although it is around 2GB/tile usually.
ncores: int
Number of cores used in multithreaded processing for orthrectification step
overwrite: bool
If True, it will overwrite existing files.
verbose: bool
Returns
-------
list
Processed files.
"""
# rajouter args:
# - outdir
# - option directory in main sinon liste de fichier .SAFE
#
# avant:
# si dir find files
# download SRTM30m
# pour chaque tiff:
# clip, calibrate, ortho
# avec l'ensemble:
# etxractROI
# multi-temp filtering
# TODO: check if temporary files are used during process, otherwise insert temporary file in processing steps
# set environment variables
os.environ['GDAL_NUM_THREADS'] = str(ncores)
os.environ['OPJ_NUM_THREADS'] = str(ncores)
os.environ['ITK_GLOBAL_DEFAULT_NUMBER_OF_THREADS'] = str(ncores)
dem_dir = dem
outdir = Path(outdir)
if orthofit:
all_steps = ['calib', 'ortho', 'mosaic', 'filter']
else:
all_steps = ['clip', 'calib', 'ortho', 'mosaic', 'filter']
if isinstance(step, str):
step = [step]
if 'all' in step:
step = all_steps
# set output directories
clip_dir = outdir / 'clipped'
if calib_dir is None:
calib_dir = outdir / 'calib'
else:
if 'clip' in step:
warnings.warn('Argument "calib_dir" should not be used with step "clip".')
calib_dir = Path(calib_dir)
ortho_dir = outdir / 'ortho'
mosaic_dir = outdir / 'mosaic'
filtered_dir = outdir / 'filtered'
if 'calib' in step:
if len(indir) == 1:
indir = indir[0]
if isinstance(indir, str):
if indir.endswith('.SAFE'):
safe_files = [Path(indir)]
else:
indir = Path(indir)
safe_files = indir.glob('*.SAFE')
elif isinstance(indir, list):
safe_files = [Path(f) for f in indir if Path(f).exists() and f.endswith('.SAFE')]
else:
raise IOError('Input directory should either be a .SAFE file (list) or be a directory containing .SAFE files.')
# safe_files = safe_files[1:5]
if direction != 'both':
safe_files_asc, safe_files_des = separateAscDesc(safe_files)
if direction == 'ascending':
safe_files = safe_files_asc
elif direction == 'descending':
safe_files = safe_files_des
else:
sys.exit('Invalid direction parameter.')
if satellite != 'both':
safe_files_s1a, safe_files_s1b = separateS1AS1B(safe_files)
if satellite == 's1a':
safe_files = safe_files_s1a
elif satellite == 's1b':
safe_files = safe_files_s1b
else:
sys.exit('Invalid satellite parameter.')
if len(safe_files) == 0:
raise IOError('No .SAFE file found in {}'.format(indir))
proc_files = []
for f in safe_files:
proc_files.extend(f.glob('measurement/*.tiff')) # files to be processed
else:
indir = Path(indir)
proc_files = indir.glob('s1*.tiff')
if (dem_dir is None) and (roi is None):
raise IOError('Without ROI, DEM directory must be given in arguments.')
if epsg is None:
if roi is None:
raise IOError('Without ROI, EPSG must be given explicitly.')
epsg = raster_epsg(roi)
#### Start processing
outdir.mkdir_p()
if (roi is not None) and any([s in ['clip', 'ortho'] for s in step]):
#roi_buffered_vector = outdir / 'buffered_roi.sqlite'
roi_buffered_vector = outdir / 'buffered_roi.geojson'
if not roi_buffered_vector.exists():
#raster_bbox(roi, buf=buffer, outfile=roi_buffered_vector)
getBufferedGeographicExtent(roi, buf=buffer, toFile=roi_buffered_vector, drvname='GeoJSON')
if (roi is not None and 'clip' in step and 'calib' in step and 'ortho' in step):
#### pipelined calibration + ortho + superimpose
ortho_dir.mkdir_p()
if dem_dir is None:
dem_dir = outdir / 'dem'
dem_dir.mkdir_p()
srtm_file = dem_dir / 'srtm.tiff'
download_srtm1(roi_buffered_vector, srtm_file, cache_dir)
if geoid is None:
geoid = download_geoid(cache_dir, verbose)
outfiles = [ortho_dir / infile.name.replace('.tiff', '_clipped_cal' + lut + '_ortho.tiff') for infile in proc_files]
mp_args = [(infile, outfile, roi, dem_dir, geoid, lut, overwrite, ram) for infile, outfile in
zip(proc_files, outfiles)]
"""
files = []
for arg in mp_args:
try:
f = S1CalibOrthoWithROI(*arg)
if f is not None:
files.append(f)
except Exception as e:
print(e)
print('Clip went wrong, file skipped: {}'.format(f))
"""
with mp.Pool(ncores) as pool:
files = pool.starmap(S1CalibOrthoWithROI, mp_args) # 1h for 8 tiles in parallel (8 threads)
proc_files = [x for x in files if x is not None]
else:
if (roi is not None and 'clip' in step):
### Clipping
if verbose:
print('Clipping process...')
clip_dir.mkdir_p()
# roi_buffered_vector = tempfile.NamedTemporaryFile(dir=outdir, suffix='.sqlite').name
# getBufferedGeographicExtent(roi, buf=50, toFile=roi_buffered_vector)
# infile, outfile, roi_buffered_vector, overwrite
outfiles = [clip_dir / infile.name.replace('.tiff', '_clipped.tiff') for infile in proc_files]
mp_args = [(infile, outfile, roi_buffered_vector, overwrite, verbose, ram) for infile, outfile in
zip(proc_files, outfiles)]
files = []
# if ncores > 1:
# with mp.Pool(ncores) as pool:
# proc_files = pool.starmap(preClip, mp_args)
# else:
for arg in mp_args:
try:
f = preClip(*arg) # 3s/S2tile
if f is not None:
files.append(f)
except Exception as e:
print(e)
print('Clip went wrong, file skipped: {}'.format(f))
proc_files = files
#### calibration
if 'calib' in step:
if verbose:
print('Calibration process...')
calib_dir.mkdir_p()
outfiles = [calib_dir / infile.name.replace('.tiff', '_cal' + lut + '.tiff') for infile in proc_files]
mp_args = [(infile, outfile, lut, overwrite, verbose, ram) for infile, outfile in zip(proc_files, outfiles)]
files = []
for arg in mp_args:
f = S1Calibration(*arg) # 6s/S2tile using 32threads
files.append(f)
proc_files = files
# with mp.Pool(ncores) as pool:
# proc_files = pool.starmap(S1Calibration, mp_args)
#### orthorectification
if 'ortho' in step:
if verbose:
print('Orthorectification process...')
ortho_dir.mkdir_p()
# download SRTM
if dem_dir is None:
dem_dir = outdir / 'dem'
dem_dir.mkdir_p()
srtm_file = dem_dir / 'srtm.tiff'
download_srtm1(roi_buffered_vector, srtm_file, cache_dir)
if geoid is None:
geoid = download_geoid(cache_dir, verbose)
outfiles = [ortho_dir / infile.name.replace('.tiff', '_ortho.tiff') for infile in proc_files]
mp_args = [(infile, outfile, epsg, dem_dir, geoid, roi, orthofit, overwrite, verbose, ram) for infile, outfile in
zip(proc_files, outfiles)]
# files = []
# for arg in mp_args:
# f = S1Ortho(*arg) #40 min/tile
# files.append(f)
# proc_files = files
with mp.Pool(ncores) as pool:
proc_files = pool.starmap(S1Ortho, mp_args) # 1h for 8 tiles in parallel (8 threads)
#### mosaic
if 'mosaic' in step:
mosaic_dir.mkdir_p()
roiname = Path(roi).name.splitext()[0]
proc_files = extractROI(proc_files, mosaic_dir, roi, roiname, ncores, overwrite, verbose, ram)
if 'superimpose' in step:
superimpose_dir = outdir / 'superimpose'
superimpose_dir.mkdir()
outfiles = [superimpose_dir / infile.name.replace('.tiff', '_superimpose.tiff') for infile in proc_files]
mp_args = [(infile, roi, outfile, overwrite, verbose, ram) for infile, outfile in zip(proc_files, outfiles)]
with mp.Pool(ncores) as pool:
proc_files = pool.starmap(Superimpose, mp_args)
#### speckle filtering
if 'filter' in step:
print(proc_files)
filtered_dir.mkdir_p()
multitempSpeckleFilt(proc_files, filtered_dir, window_radius=3, overwrite=overwrite, verbose=verbose, ram=ram)
if features is not None:
if 'filter' not in step:
print('Features only enabled if filtering is performed.')
sys.exit(-1)
feat_list = [f for f in features if f != 'vh_vv_only']
if len(feat_list) == 0:
feat_list = None
feature_dir = outdir / 'features'
feature_dir.mkdir_p()
createS1Features(filtered_dir.strip(), out_fld=feature_dir.strip(), features=feat_list,
convert_to_db=not skip_db_conversion, by_date=features_by_date)
def parse_cmd_line(argv=sys.argv[1:]):
import argparse
parser = argparse.ArgumentParser(prog='s1process',
description='Process S1 SAFE images: (clip to ROI extent ->) calibrate -> orthorectify -> filter speckle')
parser.add_argument("indir", nargs='+', help="Input directory containing .SAFE directories to be processed. It can also be a list of .SAFE directories.")
parser.add_argument('-o', '--outdir', help='Output directory where will be stored processed data.')
parser.add_argument("--roi", help="Region Of Interest raster file. If not specified, EPSG and DEM are expected.")
parser.add_argument("--epsg", type=int, help="Input EPSG. If not given, EPSG is deduced from ROI.")
parser.add_argument("--dem", help="DEM directory. Optional, if not given SRTM30m will be downloaded over ROI.")
parser.add_argument("--geoid", help="Geoid file. If not given, emg96.grd from OTB repo is used.")
parser.add_argument("--orthofit", action='store_true', help="If activated it will use mode orthofit with ROI in OTB OrthoRectification app."
"Otherwise image is clipped before calibration step and "
"aligned (OTB Superimpose) on ROI grid after mosaic")
parser.add_argument("--lut",
help='Specify lookup table to use for calibration (sigma, gamma, beta).',
choices=['sigma', 'gamma', 'beta'],
default='sigma')
parser.add_argument("--direction",
help="Direction of acquisition, either ascending, descending",
choices=['descending', 'ascending', 'both'],
default='descending')
parser.add_argument("--satellite",
help="Choose a single satellite (s1a or s1b).",
choices=['s1a', 's1b', 'both'],
default='both')
parser.add_argument('--step', nargs='+', help='Specify one or more processing steps.',
choices=['all', 'clip', 'calib', 'ortho', 'filter'], default='all')
parser.add_argument('--features', nargs='+', help="Enable the production of collected vh/vv + feature stack(supp. ratio).",
choices=['vh_vv_only', 'ratio'], default=None)
parser.add_argument('--skip_db_conversion', action='store_true', help="When feature stack production is enabled, this allows to skip"
"conversion of data to dB.")
parser.add_argument('--features_by_date', action='store_true',
help="When feature stack production is enabled, default stack compilation is"
"per-feature (vh,vv,ratios). This enables per-date stack production.")
parser.add_argument('--ncores', type=int, help='Number of cores used in multithreaded processing.', default=1)
parser.add_argument('--ram', type=int,
help='RAM limit to be used in OTB applications. Warning: ncores x ram will be used.',
default=256)
parser.add_argument('--overwrite', action='store_true', help='If True, it will overwrite existing files.')
parser.add_argument('--verbose', action='store_true', help='Verbosity')
args = parser.parse_args(argv)
return args
if __name__ == '__main__':
args = parse_cmd_line()
s1process(**vars(args))
# # to be put in option
# nproc = 12
# removing = False
# step = 'ortho'
# # Following can be ascending / descending / both
# asc_desc = 'ascending'
#
# fld = sys.argv[1]
# epsg = int(sys.argv[2])
# demfld = sys.argv[3]
# geoid = sys.argv[4]
# roi = None
# if len(sys.argv) > 5:
# roi = sys.argv[5]
# if len(sys.argv) > 6:
# asc_desc = sys.argv[6]
# if len(sys.argv) > 7 and sys.argv[7] == 'removing':
# removing = True
#
# test = glob.glob(fld + os.sep + 'measurement')
# lst = glob.glob(fld + os.sep + '*.SAFE')
#
# if len(test) > 0:
# if step == 'all':
# done = S1CalibOrtho(fld, epsg, demfld, geoid)
# elif step == 'ortho':
# done = S1Ortho(fld, epsg, demfld, geoid, removing)
# elif len(lst) > 0:
# if asc_desc != 'both':
# lst_asc, lst_des = separateAscDesc(fld)
# if asc_desc == 'ascending':
# lst = lst_asc
# elif asc_desc == 'descending':
# lst = lst_des
# else:
# sys.exit('Invalid asc/desc parameter.')
# if len(lst) > 0:
# for l in lst:
# if roi is not None:
# preClip(l, roi, removing)
# S1Calibration(l, removing, force=False)
# cmdlist = []
# for l in lst:
# cmdlist.append(['python3', 's1process.py', l, str(epsg), demfld, geoid])
# queuedProcess(cmdlist, nproc, shell=False)
# roifld = extractROI(fld, roi, nproc)
# multitempSpeckleFilt(roifld)