I try to refine orthorectification of a World-View 2  image with a set of 
GCP (17 in total). First i read .geom file in initial image with 
otbcli_ReadImageInfo and try to refine with otbcli_RefineSensorModel using 
my GCP. However, resulting orthoimage using refine .geom are more shifted 
(relative to those GCP) than original images.

Investigating this problem i found that height (z) estimate 
from otbcli_RefineSensorModel for each GCP are not the same that SRTM data. 
I use DownloadSRTMTiles application to download srtm data and the geoid 
link found on the OTB website. Here the result :

2016 Oct 17 16:29:25 : Application.logger (INFO) Adding tie point x=4222, 
y=32918, z=215.755, lon=-5.22294, lat=7.16053

2016 Oct 17 16:29:25 : Application.logger (INFO) Adding tie point x=7818, 
y=29136, z=215.046, lon=-5.20676, lat=7.17769

2016 Oct 17 16:29:25 : Application.logger (INFO) Adding tie point x=3015, 
y=30395, z=224.989, lon=-5.22845, lat=7.17191

2016 Oct 17 16:29:25 : Application.logger (INFO) Adding tie point x=2146, 
y=34611, z=227.753, lon=-5.2323, lat=7.15284

2016 Oct 17 16:29:25 : Application.logger (INFO) Adding tie point x=8825, 
y=28531, z=207.521, lon=-5.20222, lat=7.18045

2016 Oct 17 16:29:25 : Application.logger (INFO) Adding tie point x=12964, 
y=24821, z=252.431, lon=-5.18351, lat=7.19734

2016 Oct 17 16:29:25 : Application.logger (INFO) Adding tie point x=20593, 
y=20334, z=263.199, lon=-5.14906, lat=7.2178

2016 Oct 17 16:29:25 : Application.logger (INFO) Adding tie point x=19766, 
y=15039, z=259.089, lon=-5.1529, lat=7.24173

2016 Oct 17 16:29:25 : Application.logger (INFO) Adding tie point x=7572, 
y=24160, z=221.765, lon=-5.20796, lat=7.20017

2016 Oct 17 16:29:25 : Application.logger (INFO) Adding tie point x=14997, 
y=19624, z=238.779, lon=-5.17444, lat=7.22084

2016 Oct 17 16:29:25 : Application.logger (INFO) Adding tie point x=19236, 
y=23380, z=237.795, lon=-5.15517, lat=7.20398

2016 Oct 17 16:29:25 : Application.logger (INFO) Adding tie point x=20686, 
y=7548, z=259.145, lon=-5.1489, lat=7.27559

2016 Oct 17 16:29:25 : Application.logger (INFO) Adding tie point x=16714, 
y=15626, z=244.571, lon=-5.16673, lat=7.23899

2016 Oct 17 16:29:25 : Application.logger (INFO) Adding tie point x=16824, 
y=22630, z=253.821, lon=-5.16608, lat=7.20731


However, comparing my GCP with the SRTM data at the same location give me 
those results.

otb elevation estimate coord_z from SRTM 
215.755027023967 188 
215.045595474306 187 
224.989075573357 195 
227.752914068815 199 
207.520907906166 178 
252.431061319575 224 
263.198903942292 234 
259.089167755502 231 
221.765367774753 193 
238.779180367023 211 
237.794649632177 209 
259.145216934172 231 
244.570667331594 216 
253.821387341010 225 

I noticed that their is a relative common 28 m offset between those 2 
elevation

Unfortunately, it seems that OTB cannot support GCP Z coordinate as has 
been reported in 
https://groups.google.com/forum/#!searchin/otb-users/gcp$20to$20sensor%7Csort:relevance/otb-users/1Qd1mwk9m7g/QHHO-jUCFAAJ

My question remains to know if difference between elevation from my GCP and 
those estimate by OTB are the source of the shift of the orthorectification.

Thanks,¸

Félix

-- 
-- 
Check the OTB FAQ at
http://www.orfeo-toolbox.org/FAQ.html

You received this message because you are subscribed to the Google
Groups "otb-users" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/otb-users?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"otb-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to