[otb-users] Re: Developing Georeferencing and Orthorectification for RASAT satellite imagery

2016-08-10 Thread Suresh Manchikanti 
Hello ..
I am also working on automated image processing system for satellite 
imagery. 
I have RPC file for the input file, reference file and downloaded SRTM Dem 
file.

I am using the matlab's phase correlation method to identify the control 
points between input and reference images.

Now, i want to generate the ortho image from these gcps+ rpc file + dem 
file.
 I look the documentation of OTB's cookbook, but could not succeed.



Please suggest/help me in this... 

Thanks in advance.



On Sunday, 19 January 2014 20:14:59 UTC+5:30, Telepinu wrote:
>
>
> Dear fellow colleagues,
> I have been developing an automated image processing system for satellite 
> imagery. 
> So far I developed band registration, and gereferencing using a base image 
> using feature points, similar to Homologous Points Extraction.
> I use OpenCV and GDAL libraries. I use SRTM DEM. 
>
> Currently I succesfully geolocate images; but I want to *orthorectify* 
> images.
> Unfortunately we do not have a sensor model.
> I want to learn that:
>
>- Using Orfeo, can I produce a generic sensor model or shall I compute 
>them for each image?
>- How can I embed RPC file computed by Generate a RPC sensor model? Is 
>is straight forward?
>
> Best regards,
> Mustafa
>

-- 
-- 
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 otb-users@googlegroups.com
To unsubscribe from this group, send email to
otb-users+unsubscr...@googlegroups.com
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 otb-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [otb-users] OTB PostgreSQL API

2016-08-10 Thread Jiří Fejfar 
Hi Guillaume,

   - about the API, I have created new thread 
   https://groups.google.com/forum/#!topic/otb-developers/Pps8mmbHe80 ... 
   where I have just put some example of how the proposed API should work... I 
   do not want to create C++ application connecting to DB through GDAL or 
   other driver, getting data, processing with OTB and upload results back. I 
   would like to call OTB processing functions directly on data stored in 
   PostGIS using PostgreSQL backend
   - about LSMS performance... I have found in [1], Table I, that 
   non-linearity in processing time vs. image size is reported there. Could 
   you be more specific, what you mean with "improve the situation with the 
   parameters exposed in the application"? I am thinking about asking new 
   question in otb-users mail list just to cover how to use LSMS step 1 on 
   large rasters, but I would like to make some test before... 
   
Thank you very much for your suggestions, J.

[1] 
http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=6849524&url=http%3A%2F%2Fieeexplore.ieee.org%2Fiel7%2F36%2F4358825%2F06849524.pdf%3Farnumber%3D6849524

Dne středa 10. srpna 2016 11:57:54 UTC+2 Guillaume Pasero napsal(a):
>
> Hi,
>
> In OTB, almost every image is opened using GDAL. If you manage to have 
> GDAL reading from a PostGIS database, it should be transparent on OTB side.
>
> More information here: 
> https://trac.osgeo.org/gdal/wiki/frmts_wtkraster.html 
> 
> The same goes for the shape->PostGIS, if you can directly output the 
> vector segmentation in a PostGIS table, it should be faster. (
> http://www.gdal.org/drv_pg.html)
>
> About the LSMS Smoothing performance, there may be a way to improve the 
> situation with the parameters exposed in the application.
>
> Regards,
> Guillaume
>
> On 07/28/2016 09:11 AM, Jiří Fejfar wrote:
>
> Hi all,
>
> I am performing segmentation on many Landsat8 scenes stored in PostgreSQL 
> / PostGIS. I have a very sub-optimal work-flow (all is done in Python 
> script on server side):
>
>- export tiled raster from PostGIS to .tiff image 
>- perform 4 steps of LSMS 
>
> 
>  
>segmentation using python API resulting in .shp with segments 
>- import this with shp2pgsql back to table in PostGIS. 
>
> Although this is working it has several drawbacks
>
>- I have found that first step of LSMS 
>
> 
>  
>is not linear regarding elapsed time and amount of data (see attached 
>image) ... I have found not possible to process with this step whole 
>Landsat8 scene, so I am doing whole before-mentioned pipeline in cycle 
>processing 5 rows (1 consists from tiles of the size 128 x 128 px) loosing 
>advantage of artifacts free result. 
>
>
>
> 
>
>- I believe it can be much faster. 
>
>
> Because of that I am trying to find the way, how to pass tiled image from 
> PostgreSQL directly to OTB... I have found 
> http://wiki.orfeo-toolbox.org/index.php/OTB-PostGIS_Interface but it 
> seems to be focused opposite way: expose PostGIS functionality to OTB. I 
> would like to expose OTB functionality (actually only LSMS 4 steps) 
> directly to PostGIS.
>
>
> Do you know such existing API?
>
>
> If there is no such API, I would like to contribute to that. I have 
> experience creating API exposing C++ classes / functions to PostgreSQL like 
> described in 35.9.13. Using C++ for Extensibility. 
> 
>
>
> Best regards, Jiří.
> -- 
> -- 
> 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 otb-...@googlegroups.com 
> 
> To unsubscribe from this group, send email to
> otb-users+...@googlegroups.com 
> 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 otb-users+...@googlegroups.com .
> For more options, visit https://groups.google.com/d/optout.
>
>
> -- 
>  *Guillaume PASERO*
> Responsable technique
> *Business Unit E-SPACE & Geo Information - Département Image & 
> Applications*
>
> *CS Systèmes d'Information*
> Parc de la Grande Plaine - 5, Rue Brindejonc des Moulinais - BP 15872
> 31506 Toulouse Cedex 05 - FRANCE
> +33 561 17 64 21 - guillaum...@c-s.fr  
>

-- 
-- 
Check the OTB FAQ at
http://www.orfeo-t

Re: [otb-users] OTB PostgreSQL API

2016-08-10 Thread Guillaume Pasero 

  
  
Hi,
In OTB, almost every image is opened using GDAL. If you manage to
  have GDAL reading from a PostGIS database, it should be
  transparent on OTB side.
More information here: https://trac.osgeo.org/gdal/wiki/frmts_wtkraster.html

The same goes for the shape->PostGIS, if you can directly output
the vector segmentation in a PostGIS table, it should be faster. (http://www.gdal.org/drv_pg.html)

About the LSMS Smoothing performance, there may be a way to improve
the situation with the parameters exposed in the application.

Regards,
Guillaume

On 07/28/2016 09:11 AM, Jiří Fejfar
  wrote:


  Hi all,

I am performing segmentation on many Landsat8 scenes stored in
PostgreSQL / PostGIS. I have a very sub-optimal work-flow (all
is done in Python script on server side):

  export tiled raster from PostGIS to .tiff image
  perform 4 steps of LSMS
segmentation using python API resulting in .shp with
segments
  import this with shp2pgsql back to table in PostGIS.

Although this is working it has several drawbacks

  I have found that first
  step of LSMS is not linear regarding elapsed time and
amount of data (see attached image) ... I have found not
possible to process with this step whole Landsat8 scene, so
I am doing whole before-mentioned pipeline in cycle
processing 5 rows (1 consists from tiles of the size 128 x
128 px) loosing advantage of artifacts free result.





  I believe it can be much faster.



Because of that I am trying to find the way, how to pass
  tiled image from PostgreSQL directly to OTB... I have found
  http://wiki.orfeo-toolbox.org/index.php/OTB-PostGIS_Interface
  but it seems to be focused opposite way: expose PostGIS
  functionality to OTB. I would like to expose OTB functionality
  (actually only LSMS 4 steps) directly to PostGIS.


Do you know such existing API?



If there is no such API, I would like to contribute to that.
  I have experience creating API exposing C++ classes /
  functions to PostgreSQL like described in 35.9.13.
Using C++ for Extensibility.


Best regards, Jiří.

  
  -- 
  -- 
  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 otb-users@googlegroups.com
  To unsubscribe from this group, send email to
  otb-users+unsubscr...@googlegroups.com
  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 otb-users+unsubscr...@googlegroups.com.
  For more options, visit https://groups.google.com/d/optout.


-- 
  

  
  
  Guillaume PASERO
Responsable technique
Business Unit E-SPACE & Geo Information -
  Département Image & Applications

CS Systèmes d'Information
Parc de la Grande Plaine - 5, Rue Brindejonc des
Moulinais - BP 15872
31506 Toulouse Cedex 05 - FRANCE
+33 561 17 64 21 - guillaume.pas...@c-s.fr
  
  

  

  




-- 
-- 
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 otb-users@googlegroups.com
To unsubscribe from this group, send email to
otb-users+unsubscr...@googlegroups.com
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 otb-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [otb-users] qgis/remote sensing/OTB : needs & expectations ?

2016-08-10 Thread Guillaume Pasero 

  
  
Hi,
I use QGis occasionally and I am interested in OTB integration
  into QGis. My main expectations are on this integration :

  make it as close as possible to standalone OTB applications
with GUI : currently some OTB applications cannot be integrated
as such in QGis. They have to be split in several flavours
because the parameters hierarchy in OTB applications cannot be
reproduced in QGis. For instance there are 3 applications
"EdgeExtraction (gradient)", "EdgeExtraction (sobel)", "EdgeExtraction
(touzi)" instead of a single one in OTB.
  
  make it easy to use latest released OTB applications with an
existing QGis install.
  

Regards,
Guillaume

On 07/28/2016 02:01 PM, celati Laurent
  wrote:


  Good morning, 

As part of
  a mission project, I talk currently with a significant
  contributor to the development of qgis. We wonder about needs
  & expectations of the users and developers regarding
  remote sensing qgis interface (in particular, orfeo toolbox,
  wellknowed as OTB. 


So could
  you tell me, dear qgis and/or remote sensing software users
  and/or developpers your needs and expectations please?
  (creations and improvements) 

In advance,
  thank you to throw light for us and to give your opinion. This
  is a first and really important step in order to define and
  guide our possible contribution to the software. 

Kind
  regards.
  
  -- 
  -- 
  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 otb-users@googlegroups.com
  To unsubscribe from this group, send email to
  otb-users+unsubscr...@googlegroups.com
  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 otb-users+unsubscr...@googlegroups.com.
  For more options, visit https://groups.google.com/d/optout.


-- 
  

  
  
  Guillaume PASERO
Responsable technique
Business Unit E-SPACE & Geo Information -
  Département Image & Applications

CS Systèmes d'Information
Parc de la Grande Plaine - 5, Rue Brindejonc des
Moulinais - BP 15872
31506 Toulouse Cedex 05 - FRANCE
+33 561 17 64 21 - guillaume.pas...@c-s.fr
  
  

  

  




-- 
-- 
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 otb-users@googlegroups.com
To unsubscribe from this group, send email to
otb-users+unsubscr...@googlegroups.com
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 otb-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [otb-users] Orthorectification process and No data Pixel

2016-08-10 Thread Guillaume Pasero 

  
  
Hi,
I also think it is a no-data management issue. The
  Orthorectification application uses a specific background value
  when there is no data for the resampled position (for instance, at
  the corners of the orthorectified image). This background value is
  set as a no-data flag in the output geotiff. By default this
  background value is 0 (you can set it with the parameter
  outputs.default). It could be that some pixels in the shadow
  regions also have a null value, so they behave as no-data.
You can either change the background value, or remove the no-data
  flag from the output geotiff.
Regards,
Guillaume

On 08/09/2016 11:49 AM,
  insight.ci...@gmail.com wrote:


  Hi all,

i have a little problem problem to orthorectify my pleiade data
with Orfeo toolbox:

Orthorectification process generate some no data pixel into the
pleiade data ( I joined a jpg to illustrate this)

This is my workflow: 

- i open my bundle product (raw data)

- I do an optical calibration on both Pan and MSI
otbcli_OpticalCalibration -in %PAN% -out %OUT%PAN_TOC.TIF uint16
-level toc -milli 1 -ram 15000
otbcli_OpticalCalibration -in %MS% -out %OUT%MS_TOC.TIF uint16
-level toc -milli 1 -ram 15000

- i perform a bundle to perfect sensor
otbcli_BundleToPerfectSensor -inp %OUT%PAN_TOC.TIF -inxs
%OUT%MS_TOC.TIF -mode phr -out %OUT%BUNDLE.TIF uint16 -ram 15000
-elev.geoid %GEOID% -elev.dem %DEM4326% 

At this step my product is ok, with good radiometric values...
and then i try to orthorectify

-Orthorectification:
 otbcli_OrthoRecTIFication -io.in
"%OUT%BUNDLE.TIF?&skipcarto=true" -io.out
%OUT%BUNDLE_ORTHO.TIF uint16 -interpolator bco
-interpolator.bco.radius 4 -outputs.mode auto -opt.ram 15000
-elev.dem %DEM3163% -elev.geoid %GEOID% -map.epsg.code %EPSG%

My elevation data is ok, and i also try with SRTM data, but i
had the same result...
I think it's a problem with no data management, because all the
no data pixel are situated in the shadowed part of hillshade.

If someone have an idea to help me...

Thank you very much
  
  -- 
  -- 
  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 otb-users@googlegroups.com
  To unsubscribe from this group, send email to
  otb-users+unsubscr...@googlegroups.com
  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 otb-users+unsubscr...@googlegroups.com.
  For more options, visit https://groups.google.com/d/optout.


-- 
  

  
  
  Guillaume PASERO
Responsable technique
Business Unit E-SPACE & Geo Information -
  Département Image & Applications

CS Systèmes d'Information
Parc de la Grande Plaine - 5, Rue Brindejonc des
Moulinais - BP 15872
31506 Toulouse Cedex 05 - FRANCE
+33 561 17 64 21 - guillaume.pas...@c-s.fr
  
  

  

  




-- 
-- 
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 otb-users@googlegroups.com
To unsubscribe from this group, send email to
otb-users+unsubscr...@googlegroups.com
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 otb-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [otb-users] range radius and spectral radius vs. range radius and spatial radius

2016-08-10 Thread jboswellt via otb-users 
thanks!

Am Montag, 8. August 2016 13:18:24 UTC+2 schrieb Julien Michel:
>
> Dear John, 
>
> Usually range and spectral radius are the same. There must be an error 
> in the published work. 
>
> Regards, 
>
> Julien 
>
> Le 07/08/2016 à 18:39, jboswellt via otb-users a écrit : 
> > Hi all, 
> > 
> > I have a question about the terms used for the parameters hr and hs of 
> > the mean shift implementation in the otb (specifically lsms). 
> > In the tool, hr is called "range radius" and hs "spatial radius" while 
> > in this study 
> > 
> https://www.researchgate.net/profile/Simon_Rougier/publication/278033444_Improvements_of_Urban_vegetation_segmentation_and_classification_using_multi-temporal_Pleiades_images/links/557ad7fa08aee5c46044917a.pdf
>  
> > only range radius and spectral radius are mentioned. I cannot find out 
> > which one corresponds to hr and which to hs. 
> > 
> > Thanks in advance for your help! 
> > 
> > kind regards, 
> > John 
> > -- 
> > -- 
> > 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 otb-...@googlegroups.com 
>  
> > To unsubscribe from this group, send email to 
> > otb-users+...@googlegroups.com  
> > 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 otb-users+...@googlegroups.com  
> > . 
> > For more options, visit https://groups.google.com/d/optout. 
>
>
> -- 
> Julien MICHEL 
> CNES - DCT/SI/AP - BPI 1219 
> 18, avenue Edouard Belin 
> 31401 Toulouse Cedex 09 - France 
> Tel: +33 561 282 894 - Fax: +33 561 283 109 
>
>

-- 
-- 
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 otb-users@googlegroups.com
To unsubscribe from this group, send email to
otb-users+unsubscr...@googlegroups.com
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 otb-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.