[Radiance-general] about wxfalsecolor problem

[Thomas Bleicher]

Mehmedalp

Since we're a bit out of ideas here could you post the output of the getinfo
command for both pictures. That might help us understand what happened to
the images.

After reading your initial email again I found one other possible reason:

wxfalsecolor ignores pixels with red value of 0. Due to the limitations of
the RGBE image format a value of 0 in one channel means a black pixel. In
real live black pixels do not exist and are therefore ignored when an
average is calculated. This was done to allow a rectangular selection over a
masked image area without the mask contributing to the average.

You can see if a pixel contributes by moving the mouse over the image after
you have loaded the data (happens automatically for small images). If you
see an r,g,b value next to pixel position in the status bar the pixel does
count. If it's a black pixel it will show only the position.

So if your averaged area contains black regions these regions will be
ignored by wxfalsecolor but will count as 0 in ximage or Photosphere. This
might account for wxfalsecolor reporting a larger value than Photosphere.


I really should document these things someday ...

Regards,
Thomas



On Mon, Dec 13, 2010 at 2:56 PM, Greg Ward <gregoryjward at gmail.com> wrote:

> Hi Thomas,
>
> The -o option should give you original values, provided as you say
> something hasn't happened in between to lose the calibration.  I'm just
> throwing out guesses.  I have no idea why the results would disagree,
> otherwise.
>
> Cheers,
> -Greg
>
> > From: Thomas Bleicher <tbleicher at googlemail.com>
> > Date: December 13, 2010 11:14:16 AM PST
> >
> > Hi Greg.
> >
> > On Mon, Dec 13, 2010 at 12:19 PM, Greg Ward <gregoryjward at gmail.com>
> wrote:
> > I am not familiar with wxfalsecolor.  Is it possible that it is not
> taking into account the EXPOSURE setting in the header?
> >
> > As I wrote, I just use "pvalue -o ..." to get the values from the image.
> > It works fine with rendered images and test images I have created
> > with pcomb. I have never used it on HDR images, though.
> >
> > I wanted to include a feature that checks the header information
> > and warns if the images does not contain useable luminance values
> > (after pcond has been used for example). I'm not sure if I understand
> > these limitations myself correctly so I rely on the user to provide
> > appropriate images.
> >
> > Regards,
> > Thomas
>
> _______________________________________________
> Radiance-general mailing list
> Radiance-general at radiance-online.org
> http://www.radiance-online.org/mailman/listinfo/radiance-general
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.radiance-online.org/pipermail/radiance-general/attachments/20101215/4431556e/attachment.html>