Difference between revisions of "Fulla"

From PanoTools.org Wiki
Jump to navigation Jump to search
(→‎Status: digiKam)
(Undo revision 15917 by MhOOh (talk) because of distorted text)
 
(6 intermediate revisions by 2 users not shown)
Line 1: Line 1:
{{Glossary|a command line tool to batch correct [[Chromatic aberration]], [[Lens correction model|Lens distortion]] and [[Vignetting]]}}'''fulla''' is a command-line tool distributed with [[hugin]] 0.6 and later. It can correct chromatic aberration, lens distortion, and vignetting.
+
{{Glossary|a command line tool to batch correct [[Chromatic aberration]], [[Lens correction model|Lens distortion]] and [[Vignetting]]}}'''fulla''' is a command-line tool distributed with [[hugin]]. It can correct chromatic aberration, lens distortion, and vignetting.
  
 
== About fulla ==
 
== About fulla ==
Line 12: Line 12:
 
== Command-line usage ==
 
== Command-line usage ==
  
  Usage: fulla [options] inputfile(s)
+
Usage:  
  option are:
 
      -g a:b:c:d      Radial distortion coefficient for all channels, (a, b, c, d)
 
      -b a:b:c:d      Radial distortion coefficients for blue channel, (a, b, c, d)
 
                        this is applied on top of the -g distortion coefficients,
 
                        use for TCA corr
 
      -r a:b:c:d      Radial distortion coefficients for red channel, (a, b, c, d)
 
                        this is applied on top of the -g distortion coefficients,
 
                        use for TCA corr
 
      -p              Try to read radial distortion coefficients for green
 
                        channel from PTLens database
 
      -m Canon        Camera manufacturer, for PTLens database query
 
      -n Camera        Camera name, for PTLens database query
 
      -l Lens          Lens name, for PTLens database query
 
                        if not specified, a list of possible lenses is displayed
 
      -d 50            specify focal length in mm, for PTLens database query
 
      -s              do not rescale the image to avoid black borders.
 
      -f filename      Vignetting correction by flatfield division
 
                        I = I / c,    c = flatfield / mean(flatfield)
 
      -c a:b:c:d      radial vignetting correction by division:
 
                        I = I / c,    c = a + b*r^2 + c*r^4 + d*r^6
 
      -a              Correct vignetting by addition, rather than by division
 
                        I = I + c
 
      -i value        gamma of input data. used for gamma correction
 
                        before and after flatfield correction
 
      -t n            Number of threads that should be used during processing
 
      -h              Display help (this text)
 
      -o name          set output filename. If more than one image is given,
 
                        the name will be uses as suffix (default suffix: _corr)
 
      -v              Verbose
 
  
== Status ==
+
  fulla [options] inputfile(s)
As of Hugin 2010.0.0, vignetting correction seems to be [http://sourceforge.net/tracker/?func=detail&aid=3057742&group_id=77506&atid=550441 broken]. Resizing during correction is [http://sourceforge.net/tracker/?func=detail&aid=3057758&group_id=77506&atid=550444 not available] - this may cause important image parts to get cropped. Few changes have been committed to fulla since the beginning of 2008. If you want to correct individual images, best use a [http://lensfun.berlios.de/ lensfun]-based program (e.g., [http://ufraw.sourceforge.net/ UFRaw] or [http://www.digikam.org/drupal/node/319 digiKam]) these days. Lensfun also comes with a more up to date lens correction parameters database.
+
 
 +
 
 +
The following options can be specified:
 +
 
 +
* <tt>--green=db|a:b:c:d</tt> or <tt>-g db|a:b:c:d</tt>
 +
Correct radial distortion for all channels. Specify 'db' for database lookup or the 4 coefficients a:b:c:d.
 +
* <tt>--blue=db|a:b:c:d</tt> or <tt>-b db|a:b:c:d</tt>
 +
Correct radial distortion for blue channel, this is applied on top of the <tt>--green</tt> distortion coefficients, use for TCA corr. Specify 'db' for database lookup or the 4 coefficients a:b:c:d.
 +
* <tt>--red=db|a:b:c:d</tt> or <tt>-r db|a:b:c:d</tt>
 +
Correct radial distortion for red channel, this is applied on top of the <tt>--green</tt> distortion coefficients, use for TCA corr. Specify 'db' for database lookup or the 4 coefficients a:b:c:d
 +
* <tt>--camera-maker=Maker</tt> or <tt>-m Maker</tt>
 +
Camera manufacturer, for database query
 +
* <tt>--camera-model=Cam</tt> or <tt>-n Model</tt>
 +
Camera name, for database query
 +
* <tt>--lensname=Lens</tt> or <tt>-l Lens</tt>
 +
Lens name, for database query. Specify <tt>--camera-maker</tt> and <tt>--camera-model</tt> for fixed lens cameras or <tt>--lensname</tt> for interchangeable lenses.
 +
* <tt>--focallength=50</tt> or <tt>-d 50</tt>
 +
Specify focal length in mm, for database query
 +
* <tt>--aperture=3.5</tt> or <tt>-a 3.5</tt>
 +
Specify aperture for vignetting data database query
 +
* <tt>--dont-rescale</tt> or <tt>-s</tt>
 +
Do not rescale the image to avoid black borders.
 +
* <tt>--flatfield=filename</tt> or <tt>-f filename</tt>
 +
Vignetting correction by flatfield division: I = I / c, c = flatfield / mean(flatfield)
 +
* <tt>--vignetting=db|a:b:c:d</tt> or <tt>-c db|a:b:c:d</tt>
 +
Correct vignetting (by division). Specify db for database look up or the the 4 coefficients a:b:c:d: I = I / ( a + b*r^2 + c*r^4 + d*r^6)
 +
* <tt>--linear</tt>
 +
Do vignetting correction in linear color space
 +
* <tt>--gamma=value</tt> or <tt>-i value</tt>
 +
Gamma of input data. used for gamma correction before and after flatfield correction
 +
* <tt>--help</tt> or <tt>-h</tt>
 +
Display help
 +
* <tt>--output=name</tt> or <tt>-o name</tt>
 +
Set output filename. If more than one image is given, the name will be uses as suffix (default suffix: _corr)
 +
* <tt>--compression=value</tt> or <tt>-e value</tt>
 +
Compression of the output files. For jpeg output: 0-100, For tiff output: PACKBITS, DEFLATE, LZW
 +
* <tt>--offset=X:Y</tt> or <tt>-x X:Y</tt>
 +
Horizontal and vertical shift
 +
* <tt>--verbose</tt> or <tt>-v</tt>
 +
Verbose output
  
 
== Notes ==
 
== Notes ==
Line 57: Line 67:
 
* Edit ''properties''
 
* Edit ''properties''
 
* Change the ''target'' to include your correction parameters:
 
* Change the ''target'' to include your correction parameters:
   "C:\Program Files\hugin\fulla.exe" -r 0:0:0.00056:1.00015 -b 0:0:-0.00026:1.0011 -c 1:0.20:-0.51:0.2
+
   "C:\Program Files\hugin\fulla.exe" --red=0:0:0.00056:1.00015 --blue=0:0:-0.00026:1.0011 --vignetting=1:0.20:-0.51:0.2
 
* Rename the shortcut to something descriptive like ''fulla-peleng''
 
* Rename the shortcut to something descriptive like ''fulla-peleng''
 
* Now you can correct photos by dragging and dropping them onto the icon
 
* Now you can correct photos by dragging and dropping them onto the icon
 
Expanding on the above example is to create a batchfile in the hugin directory called "fulla.bat" containing the line:
 
FOR %%I IN (%1) DO C:\Program Files\hugin\fulla.exe -p -v -l Standard %%1
 
Assuming the name of the lens in the PTLens Database (see below) is "Standard" (as most are). Now, upon calling from any directory containing pics you can run "C:\Program Files\hugin\fulla.bat *.jpg" having fulla correcting all pics using the parameters retrieved from the PTLens Database mentioned below (don't forget to SET PTLENS_PROFILE)
 
 
== PTLens database lookup ==
 
 
fulla will look-up your camera in a [[PTLens]] database and try to automatically determine [[Lens correction model|Lens correction parameters]] for your pictures. An older free version of the database is required which can be downloaded from sourceforge (see [[#External links|external links]]).
 
 
Use the PTLENS_PROFILE environment variable to specify the location of the PTLens database, for example on Linux or OS X:
 
 
  PTLENS_PROFILE=$HOME/.ptlens/profile.txt
 
 
On Windows something like this should work for a PTLens standard install:
 
 
  set PTLENS_PROFILE=%ProgramFiles%\PTLens\profile.txt
 
 
otherwise enter the full path.
 
  
 
== External links ==
 
== External links ==

Latest revision as of 18:23, 3 May 2017

fulla is a command-line tool distributed with hugin. It can correct chromatic aberration, lens distortion, and vignetting.

About fulla

fulla is suitable for batch correction of:

Chromatic aberration
See tutorial in external links.
Lens distortion
Barrel or pincushion distortion can be corrected using the same a, b, c and d parameters as other Panorama Tools derived programs.
Vignetting
Polynomial correction by division, addition or flatfield images. hugin contains a vignetting calculator that can use a series of overlapping photographs to determine the likely polynomial required to correct common vignetting problems.

fulla can work with the same range of image types as nona, including 8bit and 16bit TIFF and 32bit floating-point HDR TIFF data.

Command-line usage

Usage:

  fulla [options] inputfile(s)


The following options can be specified:

  • --green=db|a:b:c:d or -g db|a:b:c:d

Correct radial distortion for all channels. Specify 'db' for database lookup or the 4 coefficients a:b:c:d.

  • --blue=db|a:b:c:d or -b db|a:b:c:d

Correct radial distortion for blue channel, this is applied on top of the --green distortion coefficients, use for TCA corr. Specify 'db' for database lookup or the 4 coefficients a:b:c:d.

  • --red=db|a:b:c:d or -r db|a:b:c:d

Correct radial distortion for red channel, this is applied on top of the --green distortion coefficients, use for TCA corr. Specify 'db' for database lookup or the 4 coefficients a:b:c:d

  • --camera-maker=Maker or -m Maker

Camera manufacturer, for database query

  • --camera-model=Cam or -n Model

Camera name, for database query

  • --lensname=Lens or -l Lens

Lens name, for database query. Specify --camera-maker and --camera-model for fixed lens cameras or --lensname for interchangeable lenses.

  • --focallength=50 or -d 50

Specify focal length in mm, for database query

  • --aperture=3.5 or -a 3.5

Specify aperture for vignetting data database query

  • --dont-rescale or -s

Do not rescale the image to avoid black borders.

  • --flatfield=filename or -f filename

Vignetting correction by flatfield division: I = I / c, c = flatfield / mean(flatfield)

  • --vignetting=db|a:b:c:d or -c db|a:b:c:d

Correct vignetting (by division). Specify db for database look up or the the 4 coefficients a:b:c:d: I = I / ( a + b*r^2 + c*r^4 + d*r^6)

  • --linear

Do vignetting correction in linear color space

  • --gamma=value or -i value

Gamma of input data. used for gamma correction before and after flatfield correction

  • --help or -h

Display help

  • --output=name or -o name

Set output filename. If more than one image is given, the name will be uses as suffix (default suffix: _corr)

  • --compression=value or -e value

Compression of the output files. For jpeg output: 0-100, For tiff output: PACKBITS, DEFLATE, LZW

  • --offset=X:Y or -x X:Y

Horizontal and vertical shift

  • --verbose or -v

Verbose output

Notes

Finding the correct lens correction parameters can be done using several approaches (see Lens correction model and external links for details). Note parameter d for radial distortion (default 1.0) is not available in GUI frontends - the d mentioned in frontends is typically a different parameter related to image shift.

Windows batch usage

Once you have calculated fulla parameters as described above, you can create a custom droplet for batch correcting photos like so:

  • Create a Shortcut to fulla.exe by dragging the fulla icon to another directory in Windows explorer
  • Edit properties
  • Change the target to include your correction parameters:
 "C:\Program Files\hugin\fulla.exe" --red=0:0:0.00056:1.00015 --blue=0:0:-0.00026:1.0011 --vignetting=1:0.20:-0.51:0.2
  • Rename the shortcut to something descriptive like fulla-peleng
  • Now you can correct photos by dragging and dropping them onto the icon

External links