rectifying digital images

classic Classic list List threaded Threaded
18 messages Options
Reply | Threaded
Open this post in threaded view
|

rectifying digital images

maning sambale
Hi,

I have a lot of aerial photos I need rectify.  The data came from an
incomplete project intending to acquire geospatial data for landslide
risk assessment.
The project ended prematurely.

Here are the details:
Images were acquired using this method.

Two GPS receivers were used, one attached to the laptop to facilitate
the navigation of the
airplane and the other to the camera to determine the geographic location of the
center of the photographs. The plane flew along the flight lines
indicated in the plan.
The photographs were taken along the flight line using a professional
Kodak N14 digital
SLR camera, which has a resolution of 14 megapixels. The camera was
fastened to a
mount, made from plywood and customized to accommodate the camera and the laptop
computer. The seat next to the pilot was detached to make room for the mount. A
camera door, designed specially for this purpose, replaced the normal Cessna 172
passenger door.

What do I have now:
over 800 photos
GPS point location for each photo
Quickbird image of the whole study area

According to their report:
There is difficulty in mosaicking the photographs. The
original plan was to stitch the photographs together to create a
composite image.
But the tilt in the photographs due to the movement of the aircraft
requires that
certain adjustments be made.

Any ideas on how to resolve this.  As far as I know, you need certain
camera parameters to orthorectify aerial photos
CAMERA NAME:               camera name______
        CAMERA IDENTIFICATION:     identification___
        CALIBRATED FOCAL LENGTH mm.:_________________
        POINT OF SYMMETRY (X)   mm.:_________________
        POINT OF SYMMETRY (Y)   mm.:_________________
        MAXIMUM NUMBER OF FIDUCIALS:_________________


How do I get this?  Is it OK to rectify the imges from the quickbird
satellite data?

cheers,

maning
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: rectifying digital images

Vincent Bain
Hello Maning,
Yes the actually important data concerns the camera properties. I fear
your images won't be easy to handle : I guess your camera is not
properly an aerial survey camera system, is it ? each image should show
fiducial marks on its edges, and you should have a calibration
certificate providing data such as focal length, fiducial marks position
and so on.
Let us know a bit more about the camera.


Le samedi 05 janvier 2008 à 15:24 +0800, maning sambale a écrit :

> Hi,
>
> I have a lot of aerial photos I need rectify.  The data came from an
> incomplete project intending to acquire geospatial data for landslide
> risk assessment.
> The project ended prematurely.
>
> Here are the details:
> Images were acquired using this method.
>
> Two GPS receivers were used, one attached to the laptop to facilitate
> the navigation of the
> airplane and the other to the camera to determine the geographic location of the
> center of the photographs. The plane flew along the flight lines
> indicated in the plan.
> The photographs were taken along the flight line using a professional
> Kodak N14 digital
> SLR camera, which has a resolution of 14 megapixels. The camera was
> fastened to a
> mount, made from plywood and customized to accommodate the camera and the laptop
> computer. The seat next to the pilot was detached to make room for the mount. A
> camera door, designed specially for this purpose, replaced the normal Cessna 172
> passenger door.
>
> What do I have now:
> over 800 photos
> GPS point location for each photo
> Quickbird image of the whole study area
>
> According to their report:
> There is difficulty in mosaicking the photographs. The
> original plan was to stitch the photographs together to create a
> composite image.
> But the tilt in the photographs due to the movement of the aircraft
> requires that
> certain adjustments be made.
>
> Any ideas on how to resolve this.  As far as I know, you need certain
> camera parameters to orthorectify aerial photos
> CAMERA NAME:               camera name______
> CAMERA IDENTIFICATION:     identification___
> CALIBRATED FOCAL LENGTH mm.:_________________
> POINT OF SYMMETRY (X)   mm.:_________________
> POINT OF SYMMETRY (Y)   mm.:_________________
> MAXIMUM NUMBER OF FIDUCIALS:_________________
>
>
> How do I get this?  Is it OK to rectify the imges from the quickbird
> satellite data?
>
> cheers,
>
> maning
> _______________________________________________
> grass-user mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-user
>

_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: rectifying digital images

Markus Metz
In reply to this post by maning sambale
Hi Maning,

if I understand you right, the purpose is to obtain a georeferenced
mosaic of the aerial photographs. There are several ways to get to there.
The easiest is to georeference each aerial photograph against your
Quickbird imagery. This is straightforward and well documented in the
GRASS help. Then you mosaic the photographs together, either with the
GRASS script or, if you're not happy with the results of i.image.mosaic,
use r.mapcalc and get inspiration from i.image.mosaic. This should
already give you a fairly good result.

if the camera was tilted against the ground which is apparently the case
with your aerial photographs, steep slopes and large differences in
elevation within an image might cause distortions which are not removed
by simple georectification, unless you use a brute force approach with
the orthorectified Quickbird imagery, i.e. at least 100 control points
per image and the strongest rubbersheeting method available.

Orthorectification is apparently only supported for analogue aerial
photography in GRASS, because you need to know points of symmetry and
the location of fiducial marks. Nowadays orthorectification of aerial
photography taken with a digital camera requires you to know not more
than the CCD resolution, focal length and flying height above ground,
but this applies only to some commercial software packages. It helps to
have a camera calibration data sheet, but you can get away without it.
Anyway, digital cameras are apparently not supported in i.ortho.photo.

I heard of some tools that can add fiducial marks to digital
photographs, but don't know where to get them and how they work, and I
would advise against patching like this.

Not sure if that helps, but good luck,

Markus
 

maning sambale wrote:

> Hi,
>
> I have a lot of aerial photos I need rectify.  The data came from an
> incomplete project intending to acquire geospatial data for landslide
> risk assessment.
> The project ended prematurely.
>
> Here are the details:
> Images were acquired using this method.
>
> Two GPS receivers were used, one attached to the laptop to facilitate
> the navigation of the
> airplane and the other to the camera to determine the geographic location of the
> center of the photographs. The plane flew along the flight lines
> indicated in the plan.
> The photographs were taken along the flight line using a professional
> Kodak N14 digital
> SLR camera, which has a resolution of 14 megapixels. The camera was
> fastened to a
> mount, made from plywood and customized to accommodate the camera and the laptop
> computer. The seat next to the pilot was detached to make room for the mount. A
> camera door, designed specially for this purpose, replaced the normal Cessna 172
> passenger door.
>
> What do I have now:
> over 800 photos
> GPS point location for each photo
> Quickbird image of the whole study area
>
> According to their report:
> There is difficulty in mosaicking the photographs. The
> original plan was to stitch the photographs together to create a
> composite image.
> But the tilt in the photographs due to the movement of the aircraft
> requires that
> certain adjustments be made.
>
> Any ideas on how to resolve this.  As far as I know, you need certain
> camera parameters to orthorectify aerial photos
> CAMERA NAME:               camera name______
> CAMERA IDENTIFICATION:     identification___
> CALIBRATED FOCAL LENGTH mm.:_________________
> POINT OF SYMMETRY (X)   mm.:_________________
> POINT OF SYMMETRY (Y)   mm.:_________________
> MAXIMUM NUMBER OF FIDUCIALS:_________________
>
>
> How do I get this?  Is it OK to rectify the imges from the quickbird
> satellite data?
>
> cheers,
>
> maning
> _______________________________________________
> grass-user mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-user
>
>
>  
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: rectifying digital images

Markus Neteler
In reply to this post by maning sambale
On Jan 5, 2008 8:24 AM, maning sambale <[hidden email]> wrote:
...
> The photographs were taken along the flight line using a professional
> Kodak N14 digital SLR camera, which has a resolution of 14 megapixels.

You can figure out camera parameters with 'jhead':
http://www.sentex.net/~mwandel/jhead/

...
> What do I have now:
> over 800 photos
> GPS point location for each photo
> Quickbird image of the whole study area
>
> According to their report:
> There is difficulty in mosaicking the photographs. The
> original plan was to stitch the photographs together to create a
> composite image.

Indeed, AFAIK you really need to do bundle block adjustment.
The i.ortho.photo module is restriced to single frame rectification.

I once received a script collection to use Octave and GRASS
to perform bundle block adjustment but never got that working
(still have the code, we could contact the author).

Randomly seeking the internet, I found
http://www.ics.forth.gr/~lourakis/sba/
sba: A Generic Sparse Bundle Adjustment C/C++ Package Based
on the Levenberg-Marquardt Algorithm (GPL'ed)

> But the tilt in the photographs due to the movement of the aircraft
> requires that certain adjustments be made.

Definitely.

Markus
--
Open Source Geospatial Foundation
http://www.osgeo.org/
http://www.grassbook.org/
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: rectifying digital images

Markus Neteler
In reply to this post by Markus Metz
On Jan 5, 2008 1:35 PM, Markus Metz <[hidden email]> wrote:
...
> Anyway, digital cameras are apparently not supported in i.ortho.photo.

Why this? I did so, using a handheld digital camera:

  M. Neteler, D. Grasso, I. Michelazzi, L. Miori, S. Merler, and
  C. Furlanello, 2005. An integrated toolbox for image registration, fusion
  and classification. International Journal of Geoinformatics. Special Issue
  on FOSS/GRASS 2004 & GIS-IDEAS 2004, 1(1), pp. 51-61, March 2005.
  PDF: http://ojs.info.gscc.osaka-cu.ac.jp/IJG/index.php/IJG/article/view/103/42

> I heard of some tools that can add fiducial marks to digital photographs

You can use the photo corners and calculate things yourself using
"jhead" or other tools.

Markus
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: rectifying digital images

Vincent Bain

Le samedi 05 janvier 2008 à 13:44 +0100, Markus Neteler a écrit :

> On Jan 5, 2008 1:35 PM, Markus Metz <[hidden email]> wrote:
> ...
> > Anyway, digital cameras are apparently not supported in i.ortho.photo.
>
> Why this? I did so, using a handheld digital camera:
>
>   M. Neteler, D. Grasso, I. Michelazzi, L. Miori, S. Merler, and
>   C. Furlanello, 2005. An integrated toolbox for image registration, fusion
>   and classification. International Journal of Geoinformatics. Special Issue
>   on FOSS/GRASS 2004 & GIS-IDEAS 2004, 1(1), pp. 51-61, March 2005.
>   PDF: http://ojs.info.gscc.osaka-cu.ac.jp/IJG/index.php/IJG/article/view/103/42
>
> > I heard of some tools that can add fiducial marks to digital photographs
>
> You can use the photo corners and calculate things yourself using
> "jhead" or other tools.

You mean pointing corners of the image as four fiducials, then assigning
them the dimensions of the sensor, in a coordinate system centered on
the estimated principal point ? If you do that with i.ortho.photo what
value would you give to the focal length, the one provided by the exif ?

Vincent

>
> Markus
> _______________________________________________
> grass-user mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-user
>

_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: rectifying digital images

Markus Neteler
On Jan 5, 2008 1:56 PM, Vincent BAIN <[hidden email]> wrote:

>
> Le samedi 05 janvier 2008 à 13:44 +0100, Markus Neteler a écrit :
>
> > On Jan 5, 2008 1:35 PM, Markus Metz <[hidden email]> wrote:
> > ...
> > > Anyway, digital cameras are apparently not supported in i.ortho.photo.
> >
> > Why this? I did so, using a handheld digital camera:
> >
> >   M. Neteler, D. Grasso, I. Michelazzi, L. Miori, S. Merler, and
> >   C. Furlanello, 2005. An integrated toolbox for image registration, fusion
> >   and classification. International Journal of Geoinformatics. Special Issue
> >   on FOSS/GRASS 2004 & GIS-IDEAS 2004, 1(1), pp. 51-61, March 2005.
> >   PDF: http://ojs.info.gscc.osaka-cu.ac.jp/IJG/index.php/IJG/article/view/103/42
> >
> > > I heard of some tools that can add fiducial marks to digital photographs
> >
> > You can use the photo corners and calculate things yourself using
> > "jhead" or other tools.
>
> You mean pointing corners of the image as four fiducials, then assigning
> them the dimensions of the sensor, in a coordinate system centered on
> the estimated principal point ? If you do that with i.ortho.photo what
> value would you give to the focal length, the one provided by the exif ?

OK, I have added my draft recipe here:
http://grass.gdf-hannover.de/wiki/Orthorectification_digital_camera

Feel free to improve the page.

Markus
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: rectifying digital images

Vincent Bain
Merci !
I'll try it soon...


Le samedi 05 janvier 2008 à 14:21 +0100, Markus Neteler a écrit :

> On Jan 5, 2008 1:56 PM, Vincent BAIN <[hidden email]> wrote:
> >
> > Le samedi 05 janvier 2008 à 13:44 +0100, Markus Neteler a écrit :
> >
> > > On Jan 5, 2008 1:35 PM, Markus Metz <[hidden email]> wrote:
> > > ...
> > > > Anyway, digital cameras are apparently not supported in i.ortho.photo.
> > >
> > > Why this? I did so, using a handheld digital camera:
> > >
> > >   M. Neteler, D. Grasso, I. Michelazzi, L. Miori, S. Merler, and
> > >   C. Furlanello, 2005. An integrated toolbox for image registration, fusion
> > >   and classification. International Journal of Geoinformatics. Special Issue
> > >   on FOSS/GRASS 2004 & GIS-IDEAS 2004, 1(1), pp. 51-61, March 2005.
> > >   PDF: http://ojs.info.gscc.osaka-cu.ac.jp/IJG/index.php/IJG/article/view/103/42
> > >
> > > > I heard of some tools that can add fiducial marks to digital photographs
> > >
> > > You can use the photo corners and calculate things yourself using
> > > "jhead" or other tools.
> >
> > You mean pointing corners of the image as four fiducials, then assigning
> > them the dimensions of the sensor, in a coordinate system centered on
> > the estimated principal point ? If you do that with i.ortho.photo what
> > value would you give to the focal length, the one provided by the exif ?
>
> OK, I have added my draft recipe here:
> http://grass.gdf-hannover.de/wiki/Orthorectification_digital_camera
>
> Feel free to improve the page.
>
> Markus
>

_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: rectifying digital images

maning sambale
Salamat!

Never expected such quick replies!

A few more details:
Camera:  Kodak Pro DCS 14N digital SLR camera
using a 19-35mm Tamron zoom lens set at 19mm focal length, 13.5
megapixel resolution (4,500 x 3,000 pixels).

Images are in Kodak DCR format.  Converting to tif before export to
GRASS more than double the size.

I have the following parameters for each image:
FILENAME
IMAGE NUMBER
DATE
TIME
GPS STATUS
LATITUDE
LONGITUDE
ALTITUDE
ISO SPEED
APERTURE
SHUTTER FOCAL LENGTH

> unless you use a brute force approach with
> the orthorectified Quickbird imagery, i.e. at least 100 control points
> per image and the strongest rubbersheeting method available.
I think I'd go this route as a last resort, I have approximately 800
images to rectify.


> > http://grass.gdf-hannover.de/wiki/Orthorectification_digital_camera
Thanks for this one

>>
http://ojs.info.gscc.osaka-cu.ac.jp/IJG/index.php/IJG/article/view/103/42

and this one!

We'll keep you posted as I do this.  

maning

On Sat, 2008-01-05 at 14:44 +0100, Vincent BAIN wrote:

> Merci !
> I'll try it soon...
>
>
> Le samedi 05 janvier 2008 à 14:21 +0100, Markus Neteler a écrit :
> > On Jan 5, 2008 1:56 PM, Vincent BAIN <[hidden email]> wrote:
> > >
> > > Le samedi 05 janvier 2008 à 13:44 +0100, Markus Neteler a écrit :
> > >
> > > > On Jan 5, 2008 1:35 PM, Markus Metz <[hidden email]> wrote:
> > > > ...
> > > > > Anyway, digital cameras are apparently not supported in i.ortho.photo.
> > > >
> > > > Why this? I did so, using a handheld digital camera:
> > > >
> > > >   M. Neteler, D. Grasso, I. Michelazzi, L. Miori, S. Merler, and
> > > >   C. Furlanello, 2005. An integrated toolbox for image registration, fusion
> > > >   and classification. International Journal of Geoinformatics. Special Issue
> > > >   on FOSS/GRASS 2004 & GIS-IDEAS 2004, 1(1), pp. 51-61, March 2005.
> > > >   PDF: http://ojs.info.gscc.osaka-cu.ac.jp/IJG/index.php/IJG/article/view/103/42
> > > >
> > > > > I heard of some tools that can add fiducial marks to digital photographs
> > > >
> > > > You can use the photo corners and calculate things yourself using
> > > > "jhead" or other tools.
> > >
> > > You mean pointing corners of the image as four fiducials, then assigning
> > > them the dimensions of the sensor, in a coordinate system centered on
> > > the estimated principal point ? If you do that with i.ortho.photo what
> > > value would you give to the focal length, the one provided by the exif ?
> >
> > OK, I have added my draft recipe here:
> > http://grass.gdf-hannover.de/wiki/Orthorectification_digital_camera
> >
> > Feel free to improve the page.
> >
> > Markus
> >
>
> _______________________________________________
> grass-user mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-user

_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: rectifying digital images

Vincent Bain
If you follow Markus' how-to, you may be interested in technical info
concerning your camera :

Kodak DCS Pro SLR/n/c
imager size : 36.0mm x 24.0mm
pixel size : 7.9µm x 7.9µm

see this page for other cameras :
http://www.digitaldingus.com/reference/general/sensorsizes.php

Good luck !

Vincent




Le samedi 05 janvier 2008 à 23:25 +0800, Maning Sambale a écrit :

> Salamat!
>
> Never expected such quick replies!
>
> A few more details:
> Camera:  Kodak Pro DCS 14N digital SLR camera
> using a 19-35mm Tamron zoom lens set at 19mm focal length, 13.5
> megapixel resolution (4,500 x 3,000 pixels).
>
> Images are in Kodak DCR format.  Converting to tif before export to
> GRASS more than double the size.
>
> I have the following parameters for each image:
> FILENAME
> IMAGE NUMBER
> DATE
> TIME
> GPS STATUS
> LATITUDE
> LONGITUDE
> ALTITUDE
> ISO SPEED
> APERTURE
> SHUTTER FOCAL LENGTH
>
> > unless you use a brute force approach with
> > the orthorectified Quickbird imagery, i.e. at least 100 control points
> > per image and the strongest rubbersheeting method available.
> I think I'd go this route as a last resort, I have approximately 800
> images to rectify.
>
>
> > > http://grass.gdf-hannover.de/wiki/Orthorectification_digital_camera
> Thanks for this one
>
> >>
> http://ojs.info.gscc.osaka-cu.ac.jp/IJG/index.php/IJG/article/view/103/42
>
> and this one!
>
> We'll keep you posted as I do this.  
>
> maning
>
> On Sat, 2008-01-05 at 14:44 +0100, Vincent BAIN wrote:
> > Merci !
> > I'll try it soon...
> >
> >
> > Le samedi 05 janvier 2008 à 14:21 +0100, Markus Neteler a écrit :
> > > On Jan 5, 2008 1:56 PM, Vincent BAIN <[hidden email]> wrote:
> > > >
> > > > Le samedi 05 janvier 2008 à 13:44 +0100, Markus Neteler a écrit :
> > > >
> > > > > On Jan 5, 2008 1:35 PM, Markus Metz <[hidden email]> wrote:
> > > > > ...
> > > > > > Anyway, digital cameras are apparently not supported in i.ortho.photo.
> > > > >
> > > > > Why this? I did so, using a handheld digital camera:
> > > > >
> > > > >   M. Neteler, D. Grasso, I. Michelazzi, L. Miori, S. Merler, and
> > > > >   C. Furlanello, 2005. An integrated toolbox for image registration, fusion
> > > > >   and classification. International Journal of Geoinformatics. Special Issue
> > > > >   on FOSS/GRASS 2004 & GIS-IDEAS 2004, 1(1), pp. 51-61, March 2005.
> > > > >   PDF: http://ojs.info.gscc.osaka-cu.ac.jp/IJG/index.php/IJG/article/view/103/42
> > > > >
> > > > > > I heard of some tools that can add fiducial marks to digital photographs
> > > > >
> > > > > You can use the photo corners and calculate things yourself using
> > > > > "jhead" or other tools.
> > > >
> > > > You mean pointing corners of the image as four fiducials, then assigning
> > > > them the dimensions of the sensor, in a coordinate system centered on
> > > > the estimated principal point ? If you do that with i.ortho.photo what
> > > > value would you give to the focal length, the one provided by the exif ?
> > >
> > > OK, I have added my draft recipe here:
> > > http://grass.gdf-hannover.de/wiki/Orthorectification_digital_camera
> > >
> > > Feel free to improve the page.
> > >
> > > Markus
> > >
> >
> > _______________________________________________
> > grass-user mailing list
> > [hidden email]
> > http://lists.osgeo.org/mailman/listinfo/grass-user
>
> _______________________________________________
> grass-user mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-user
>

_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: rectifying digital images

maning sambale
In reply to this post by Markus Neteler

> OK, I have added my draft recipe here:
> http://grass.gdf-hannover.de/wiki/Orthorectification_digital_camera

It says:
Usually such high-resolution DEMs (raster cell length below 1m) are not
easily available. Therefore, interpolation of a given DEM to a higher
resolution is recommended

The dem I have is the SRTM data (90m).  Eventhough I can interpolate
them I don't think it can be useful for i.ortho.photo's purpose.  Or is
it?

>>I once received a script collection to use Octave and GRASS
to perform bundle block adjustment but never got that working
(still have the code, we could contact the author).

Can we use this (SBA)?

> Markus
> _______________________________________________
> grass-user mailing list
> [hidden email]
> http://lists.osgeo.org/mailman/listinfo/grass-user

_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: rectifying digital images

sebastian sauer-2
In reply to this post by Vincent Bain
Hi,

Sat 05 Jan 2008 16:45, Vincent BAIN wrote:
> If you follow Markus' how-to, you may be interested in technical info
> concerning your camera :
>
> Kodak DCS Pro SLR/n/c
> imager size : 36.0mm x 24.0mm
> pixel size : 7.9µm x 7.9µm

exactly. even if you (for unknown reasons) do not follow Markus' how-to,
you will need this information. and it won't hurt double-checking the
correctness of this data against, the DPI in the EXIF data.

given that, it's pretty straight-forward to set-up precise fiducial
marks.

(cf. http://edclxs22.cr.usgs.gov/osl/figure2.gif &&
http://grass.itc.it/grass63/manuals/html63_user/photo.camera.html )

*but* what is tricky is getting the calibrated focal length right.
(and this of course (distortions!) is a quite important parameter)

> Le samedi 05 janvier 2008 à 23:25 +0800, Maning Sambale a écrit :
> > Camera:  Kodak Pro DCS 14N digital SLR camera
> > using a 19-35mm Tamron zoom lens set at 19mm focal length
ok, his info is not so shabby, at least you know the "calibrated focal
length" should be ~19mm.

also the spec of the lens should include the focal length tolerance.
(aka by how many percent the calibrated focal length is off from the
value of 19mm.)

cf. http://grass.itc.it/grass63/manuals/html63_user/photo.camera.html &&
and thread "[CREATE] Lens correction database" @
http://lists.freedesktop.org/archives/create/2007-May/thread.html#743

> > SHUTTER FOCAL LENGTH
?? not quite sure. are these actually to different paramenters, namely
the shutter speed && the focal lenght? or what is this??

cheers,
s.

ps: if anyone knows a good trick to do a precise focal length
calibration, plz don't keep it as your secret, but share it with us :))
SCNR.
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: rectifying digital images

Markus Neteler
In reply to this post by maning sambale
On Jan 5, 2008 4:46 PM, Maning Sambale <[hidden email]> wrote:

>
> > OK, I have added my draft recipe here:
> > http://grass.gdf-hannover.de/wiki/Orthorectification_digital_camera
>
> It says:
> Usually such high-resolution DEMs (raster cell length below 1m) are not
> easily available. Therefore, interpolation of a given DEM to a higher
> resolution is recommended
>
> The dem I have is the SRTM data (90m).  Eventhough I can interpolate
> them I don't think it can be useful for i.ortho.photo's purpose.  Or is
> it?

Well, you have what you have, I think. For which country do you need
the DEM?
Resampling helps at least to avoid ugly stairs (of course the information
content remains the same).

> >>I once received a script collection to use Octave and GRASS
> to perform bundle block adjustment but never got that working
> (still have the code, we could contact the author).
>
> Can we use this (SBA)?

Will write to the author, but my contact is from 2002.

Markus
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: rectifying digital images

Markus Neteler
Hi Man,

not sure if you remember me - we spoke about bundle block adjustment
in 2002.

There is a recent discussion about this in the GRASS user list and
I remembered that I have received "atgrass.tgz" from you. Did you
continue to work on this?

Best regards
Markus Neteler

On Jan 5, 2008 5:31 PM, Markus Neteler <[hidden email]> wrote:

> On Jan 5, 2008 4:46 PM, Maning Sambale <[hidden email]> wrote:
> > I once received a script collection to use Octave and GRASS
> > to perform bundle block adjustment but never got that working
> > (still have the code, we could contact the author).
> >
> > Can we use this (SBA)?
>
> Will write to the author, but my contact is from 2002.
>
> Markus




--
Open Source Geospatial Foundation
http://www.osgeo.org/
http://www.grassbook.org/
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: rectifying digital images

sebastian sauer-2
In reply to this post by sebastian sauer-2
Sat 05 Jan 2008 17:13, sebastian sauer wrote:

> Sat 05 Jan 2008 16:45, Vincent BAIN wrote:
> > If you follow Markus' how-to, you may be interested in technical info
> > concerning your camera :
> >
> > Kodak DCS Pro SLR/n/c
> > imager size : 36.0mm x 24.0mm
> > pixel size : 7.9µm x 7.9µm
>
> exactly. even if you (for unknown reasons) do not follow Markus' how-to,
> you will need this information. and it won't hurt double-checking the
> correctness of this data against, the DPI in the EXIF data.
>
> given that, it's pretty straight-forward to set-up precise fiducial
> marks.

turns out this is essentially known to the experts :)
cf. http://www.elcovision.com/e_kamera.html
"[DSLRs] have very low deformations; a réseau grid plate for compensation
of film or paper deformations is not necessary. For the definition of a photo
coordinate system the four photo corners are sufficient. This means:
Digital camera do not need to be modified, but can be calibrated and used
for measurement as they are."

nb.: the markings on a réseau grid plate [0] are the fiducial marks (aka réseau
marks.)

[0] http://www.elcovision.com/img/Reseauplatte.jpg

cf. http://www.ee.oulu.fi/~jth/calibr/ 
"Modern CCD cameras are usually capable of a spatial accuracy greater than
1/50 of the pixel size."

>
> (cf. http://edclxs22.cr.usgs.gov/osl/figure2.gif &&
> http://grass.itc.it/grass63/manuals/html63_user/photo.camera.html )
>
> *but* what is tricky is getting the calibrated focal length right.
> (and this of course (distortions!) is a quite important parameter)
mmh, i'm still skeptical, as i frankly really have no glue how accurate
the EXIF data really is. (hints are highly appreciated! e.g. would be
interesting if this is within the accuracy of the autofocus error of the
lens. )

but according to the experts at least decent DSLR cameras seem to produce
a quite accurate EXIF focal length data.
cf.: http://www.elcovision.com/e_elco_reseau.html

BTW, does anyone know how the EXIF focal length is defined?

> > Le samedi 05 janvier 2008 à 23:25 +0800, Maning Sambale a écrit :
> > > Camera:  Kodak Pro DCS 14N digital SLR camera
> > > using a 19-35mm Tamron zoom lens set at 19mm focal length
> ok, his info is not so shabby, at least you know the "calibrated focal
> length" should be ~19mm.
>
> also the spec of the lens should include the focal length tolerance.
seems which data is hard to get for most (D)SLR lenses.

anyway to get a "feeling" what is roughly possible for a single lens.
Focal Length Tolerance: ± 0.5 %
http://www.rainbowoptics.com/Spherical_Lenses.aspx

but of course a (D)SLR lenses is a cascade of many lenses. in
professional analgue photogammetry, Focal Length Tolerance: ± 3 mm @153mm
(so this should be ~ 4.6 % )
http://erosproject.cr.usgs.gov/osl/reqtable.html 

BTW, does anyone know what camera-sizes are usually used in aerial
photogammetry? aka how big is the plate? medium-size? large-size?

> ps: if anyone knows a good trick to do a precise focal length
> calibration, plz don't keep it as your secret, but share it with us :))
> SCNR.
http://www.google.com/search?hl=en&q=camera+calibration&btnG=Search
http://www.stoske.de/digicam/Artikel/verzeichnung.html
http://wiki.panotools.org/Lens_correction_model
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: rectifying digital images

maning sambale
In reply to this post by Markus Neteler
Markus,

> Well, you have what you have, I think. For which country do you need
> the DEM?
Philippines

> > >>I once received a script collection to use Octave and GRASS
> > to perform bundle block adjustment but never got that working
> > (still have the code, we could contact the author).
I've been reading through a couple of articles using bundle block
adjustment.  

>From the manual of Leica Photogrammetry Suite:
Bundle block adjustment determines the position and orientation for each
image in a project as they existed at the time of photographic or image
exposure. The resulting parameters are referred to as exterior
orientation parameters. In order estimate the exterior orientation
parameters, a minimum three GCPs is required for the entire block,
regardless of how many images are contained within the project.    

This maybe what GRASS needs as an additional functionality.

cheers,

maning

_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: rectifying digital images

Markus Neteler
On Jan 7, 2008 3:03 PM, Maning Sambale <[hidden email]> wrote:
> Markus,
>
> > Well, you have what you have, I think. For which country do you need
> > the DEM?
> Philippines

You could also try ASTER DEM (30m).
And ask on this mailing list:
http://www.osgeo.org/geodata

> > > >>I once received a script collection to use Octave and GRASS
> > > to perform bundle block adjustment but never got that working
> > > (still have the code, we could contact the author).
> I've been reading through a couple of articles using bundle block
> adjustment.
>
> >From the manual of Leica Photogrammetry Suite:
> Bundle block adjustment determines the position and orientation for each
> image in a project as they existed at the time of photographic or image
> exposure. The resulting parameters are referred to as exterior
> orientation parameters. In order estimate the exterior orientation
> parameters, a minimum three GCPs is required for the entire block,
> regardless of how many images are contained within the project.
>
> This maybe what GRASS needs as an additional functionality.

I have found the author's recent email address of these BBA scripts,
will le you know if he responds.

Markus
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: rectifying digital images

sebastian sauer-2
Mon 07 Jan 2008 15:45, Markus Neteler wrote:

> On Jan 7, 2008 3:03 PM, Maning Sambale <[hidden email]> wrote:
> > >From the manual of Leica Photogrammetry Suite:
> > Bundle block adjustment determines the position and orientation for each
> > image in a project as they existed at the time of photographic or image
> > exposure. The resulting parameters are referred to as exterior
> > orientation parameters. In order estimate the exterior orientation
> > parameters, a minimum three GCPs is required for the entire block,
> > regardless of how many images are contained within the project.
> >
> > This maybe what GRASS needs as an additional functionality.
Maning, c'mon what does this Leica marketing blah-blah help??

if it's *that easy* plz feel free to implement this "additional
functionality". actually there exists a pretty nice, quite well known
general purpose implementation. http://www.ics.forth.gr/~lourakis/sba/

(maybe OSSIM might have parts of this "additional functionality"
implemented.)

also somehow, i fear you just don't realize that your images have a
very high lens-distortion, while professional arial cameras produce
quite rectilinear images.


if its of general interest on this mailing-list, i can explain that in
detail.
(personally i also strongly believe lens correction should *not* become
part of the GRASS code-base, as it would a duplication of effort. and
very recently there's quite some high-quality knowledge (and even some
quite promising code) in the FOSS-community, which would be also perfect
for our photogammetric needs.)


> I have found the author's recent email address of these BBA scripts,
> will le you know if he responds.
having this code, would surely not hurt :)


cheers,
s.
_______________________________________________
grass-user mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-user