# pamsistoaglyph - convert a single-image stereogram to a red/cyan anagly... - man(1) - [phpMan] *Source: tldr-pages* --- _Pamsistoaglyph_ _User_ [_Manual_(1)](https://www.chedong.com/phpMan.php/man/Manual/1/markdown) General Commands Manual _Pamsistoaglyph_ _User_ [_Manual_(1)](https://www.chedong.com/phpMan.php/man/Manual/1/markdown) ## NAME pamsistoaglyph - convert a single-image stereogram to a red/cyan anaglyphic image ## SYNOPSIS **pamsistoaglyph **[--**invert**] [--**sep**=_number_] [--**minsep**=_number_] [--**gray**=_number_] [_in_netpbmfile_ All options can be abbreviated to their shortest unique prefix. You may use either white space or an equals sign between an option name and its value. ## DESCRIPTION This program is part of [**Netpbm**(1)](https://www.chedong.com/phpMan.php/man/Netpbm/1/markdown). **pamsistoaglyph **reads a Netpbm image as input and produces a Netpbm image as output. **pamsistoaglyph **takes a single-image stereogram (SIS) such as those produced by **pamstere‐** [**ogram**(1)](https://www.chedong.com/phpMan.php/man/ogram/1/markdown) and converts it to a red/cyan anaglyphic image such as those produced by [**ppm3d**(1)](https://www.chedong.com/phpMan.php/man/ppm3d/1/markdown). Many people have trouble tricking their eyes into focusing beyond the image in front of them and are therefore unable to perceive the 3-D shape hidden within a single-image stereogram. Anaglyphic stereograms are easier to perceive in 3-D but require a pair of red/cyan glasses such as those often used to watch 3-D movies. The goal of **pamsistoaglyph **is to help people who have trouble viewing single-image stereograms see the intriguing 3-D effect. **pamsistoaglyph **can convert single-image random-dot stereograms (SIRDS), wallpaper stere‐ ograms, and even dual-image stereograms to anaglyphic images. ## OPTIONS In addition to the options common to all programs based on libnetpbm (most notably **-quiet**, see [ Common Options](index.html#commonoptions) ), **pamsistoaglyph **recognizes the following command line options: For most images, no command-line options need to be specified. The following options are available, however, for unusual circumstances. ### --invert Swap the left- and right-eye images. **pamsistoaglyph **assumes that its input represents a wall-eyed stereogram and generates the anaglyphic image accordingly. If the generated image appears to recede into the page where it should pop out of the page (and vice versa), this typically implies that the input image represents a cross-eyed stereogram. Use **--invert **to correct the image depth. **--sep**=_number_ Specify the distance in pixels between the left- and right-eye images. Essentially, this corresponds to the distance between repetitions of the background pattern. The **--sep** option should rarely be necessary as **pamsistoaglyph **is fairly good at determining automatically the eye-separation distance. **--minsep**=_number_ This option is similar to **--sep **but constrains **pamsistoaglyph **only to a _minimum_ eye-separation distance. Any distance larger than _number_ is acceptable. The **--minsep** option should rarely be necessary as **pamsistoaglyph **is fairly good at determining automatically the eye-separation distance. The default value for the minimum eye-separation distance is 10% of the image width; this value seems to work well in practice. **--gray**=_number_ Limit the number of gray levels to use when searching for the optimal eye-separation distance. Because **pamsistoaglyph **looks for repeated patterns, it is vulnerable to being confused by slight variations in color. By reducing the input image to grayscale and capping the number of gray levels, **pamsistoaglyph **ameliorates the effects of unintentional color variations (such as those caused by conversion from a low-quality JPEG image, for example). The default of 63 seems to work well so the **--gray **option should rarely be necessary. ## NOTES The registration algorithm used by **pamsistoaglyph **was developed specifically for this pro‐ gram. As far as the author knows, there are no existing algorithms for converting stereograms to anaglyphs. The algorithm works as follows: • Convert the image to grayscale to increase the ability to identify matches. • Count the number of pixels that match _N_ pixels ahead in the image for all _N_ in [1, _width_/2]. • Maintain a running mean (mu) and standard deviation (sigma) of the number of matched pixels. • Store the _N_ corresponding to each spike in the number of matched pixels. A spike is defined as a tally that exceeds the mean plus one, two, or three standard deviations. Only the first spike of a given standard-deviation multiplier is stored. • If a tally greater than mu+3sigma was encountered, return the corresponding _N_. If not, then if a tally greater than mu+2sigma was encountered, return the corresponding _N_. If not, then if a tally greater than mu+sigma was encountered, return the corresponding _N_. If not, then return the _N_ that produces the minimum average distance between matched pixels (i.e., _#matches_ divided by _#pixels_). If no such _N_ exceeds the minimum allowable eye-separation value, return zero to indicate failure. • If the algorithm returned zero, rerun the algorithm independently on each row of the input image and return the median of all _N_ that exceed the minimum allowable eye-separation value. If no such _N_ exists, abort with an error message. ## HISTORY Scott Pakin wrote **pamsistoaglyph **in April 2009. It first appeared in Netpbm in Release 10.47 (June 2009). ## AUTHOR Copyright (C) 2009 Scott Pakin, _scott+ ## SEE ALSO • [**pamstereogram**(1)](https://www.chedong.com/phpMan.php/man/pamstereogram/1/markdown) • [**ppm3d**(1)](https://www.chedong.com/phpMan.php/man/ppm3d/1/markdown), • []() ## DOCUMENT SOURCE This manual page was generated by the Netpbm tool 'makeman' from HTML source. The master documentation is at **** netpbm documentation 05 April 2009 _Pamsistoaglyph_ _User_ [_Manual_(1)](https://www.chedong.com/phpMan.php/man/Manual/1/markdown)