Panotools::Script 0.27 released

84 views
Skip to first unread message

Bruno Postle

unread,
Jul 11, 2012, 6:40:50 PM7/11/12
to Hugin ptx
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Panotools::Script is a perl module for reading/writing and
manipulating Hugin .pto project files, this 0.27 release has some
updates and fixes some bugs:

- - A new 'ptomorph' tool, looks at control points in a Hugin PTO
panorama project and creates a new set of photos that are distorted
such that these control points line up perfectly. Distortion is a
'rubber-sheet' type morph and depends on ImageMagick. This is a
prototype tool, the idea is that a similar feature can be added to
Hugin/nona at some point.

- - A new 'ptoreset' tool to reset photometric parameters in a .pto
file, similar to the Hugin Reset function. Note that this also
implements white balance equalisation from EXIF, which isn't yet
supported by Hugin.

A Panotools::Script tarball will be available on CPAN soon:
http://search.cpan.org/dist/Panotools-Script/

0.27 changes:

remove linux desktop entries for tif2svg-gui enblend-svg process-masks-gui qtvr2erect-gui
apparently there is a bsd called dragonfly
new tool: ptoreset - reset photometric parameters in a .pto file
fix gigatile bug with uncropped output (Emad ud din Bhatt)
remove ptograph as superceded by Hugin layout mode
skip Panotools::Makefile tests if no suitable make executable found
some methods for fast querying of transformation coordinates via pano_trafo
new tool: ptomorph - distort photos before stitching
try and fix division by zero error when EXIF ExposureTime = 0 (Tobias Mueller)
erect2planet and erect2mercator now take multiple input files

SHA1SUM: c1df147be2a112824f0989fe4573b6dc9a908573 Panotools-Script-0.27.tar.gz

- --
Bruno
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (GNU/Linux)

iD8DBQFP/gDwFqOhwCjyCLoRArs/AKC2+iDT6oYJbrlwppe4sfuh4QNiDACgrObt
aLoeapJZkLswrhV625Fgz6A=
=1ypg
-----END PGP SIGNATURE-----

Tduell

unread,
Jul 13, 2012, 9:45:13 PM7/13/12
to hugi...@googlegroups.com
Hello Bruno,


On Thursday, July 12, 2012 8:40:50 AM UTC+10, Bruno Postle wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Panotools::Script is a perl module for reading/writing and
manipulating Hugin .pto project files, this 0.27 release has some
updates and fixes some bugs:

- - A new 'ptomorph' tool, looks at control points in a Hugin PTO
panorama project and creates a new set of photos that are distorted
such that these control points line up perfectly.  

Thanks for the new features.
The availability of 'ptomorph' has prompted me to revisit my work on deblurring an overlapped sharp/blur pair, which had defeated me.
I was curious to see what effect 'ptomorph' would have on the point spread function I derive from the sharp/blur pair.
Experimenting with Shepard's distortion and polynomial distortion, I find that when using the polynomial the control points are not written to the resulting pto file. Is this an oversight?

Cheers,
Terry

Bruno Postle

unread,
Jul 14, 2012, 5:21:24 PM7/14/12
to Hugin ptx
On Fri 13-Jul-2012 at 18:45 -0700, Terry Duell wrote:

> Experimenting with Shepard's distortion and polynomial distortion,
> I find that when using the polynomial the control points are not
> written to the resulting pto file. Is this an oversight?

It took a while to remember why I did this. The morph moves
features about in the source images, this is quite predictable with
the Shepard's distortion, so ptomorph rewrites the control point
locations to suit - If you open up the new project in Hugin then the
control points will be in the right place.

The position of features with the polynomial distortion is not so
predictable, as it is done in ImageMagick, so I just delete the
control points since I don't know where to put them anyway.

--
Bruno

Terry Duell

unread,
Jul 14, 2012, 7:34:08 PM7/14/12
to hugi...@googlegroups.com
Hello Bruno,
OK, I understand.
I had hoped to be able to have a closer look at how the different
distortion methods affected regions around a control point to get a better
idea of which distortion method might be best suited to deriving an
improved psf for the blur/sharp problem. Not to worry, a bit of
experimenting may provide the answer.

Cheers,
--
Regards,
Terry Duell
Reply all
Reply to author
Forward
0 new messages