[Open-graphics] Yuv->RGB conversion

[Nicolas Boulay]

I copy/past the code i use.
clip8 trunc the value at 0 or 255.

YUV444toRGB888_true are the value find inside the fourcc website and a
microsoft web site.

YUV444toRGB888 are the one i use because the array of possible value are not
the same. I find this because on some image appear big array of white or
blue.

I juste change Y value ( 298 -> 266) because i did not find an easy way to
have the max and the min for the uv part of the image.

/*
  Real formule for yuv -> rgb conversion

  "the defined range for Y is [16,235] (220 steps) and the valid ranges for
Cr
  and Cb are [16,239] (235 steps) These are normalized ranges and when the
  data is used, 16 is added to Y and 128 is added from Cr and Cb to
de-normalize
  them. The reason for this is that control and colorburst information is
stored
  in the components along with the luminence and chromiance data which is
why
  the full range of [0,255] is not used. There is no "rumor" about this
being
  correct. It is a defined standard for YCrCb video. I wouldn't be surprised
if
  certain PC programs are using a variant where the entire range of [0,255]
is
  used but technically it is not YCrCb." http://www.fourcc.org/fccyvrgb.php

  Y     [16,235] for [0.0:1.0]
  U&V   [16,239] for [-0.5;+0.5]
*/

inline unsigned char * video_t::YUV444toRGB888_true(unsigned int Y,
     unsigned int U,
     unsigned int V,
     unsigned char ret[3])
{
  int C, D, E;
  C = Y - 16;
  D = U - 128;
  E = V - 128;

  ret[0] = clip8(( 298 * C           + 409 * E + 128) >> 8);
  ret[1] = clip8(( 298 * C - 100 * D - 208 * E + 128) >> 8);
  ret[2] = clip8(( 298 * C + 516 * D           + 128) >> 8);

  return ret;
}

/*
  but webcam use Y[0;244]
*/
inline unsigned char * video_t::YUV444toRGB888(unsigned int Y,
     unsigned int U,
     unsigned int V,
     unsigned char ret[3])
{
  int C, D, E;
  C = Y;
  D = U - 128;
  E = V - 128;

  ret[0] = clip8(( 266 * C           + 409 * E + 128) >> 8);
  ret[1] = clip8(( 266 * C - 100 * D - 208 * E + 128) >> 8);
  ret[2] = clip8(( 266 * C + 516 * D           + 128) >> 8);

  return ret;
}


2007/1/19, Timothy Miller <theosib at gmail.com>:
> On 1/19/07, Nicolas Boulay <nicolas.boulay at gmail.com> wrote:
>
> > So we can't take a fixe transformation matrix for it.
>
> I think someone here said that the various transform matrices aren't
> really different enough for it to make much of a visible difference.
> If there are a handful of different ones that are really different, we
> will have to decide whether it's better to have multiple fixed
> functions versus one generalized one.  Some of the subterms may be
> shared between different sets of coefficients.
>
> --
> Timothy Miller
> http://www.cse.ohio-state.edu/~millerti
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.duskglow.com/open-graphics/attachments/20070121/46014043/attachment.html