It's hard to tell from this exactly what's going on, since I don't know exactly
how you create the ImageBuf in ip_histogram_to_image.
If you are returning an ImageBuf, then put it on the image stack (ot.push...)
and used the -o command, there should be no reason for you to call save
directly.
Also, I'm presuming that the dimensions of the histogram image actually matches
specR, otherwise you're going to really confuse it with the Rib = *histogram_R.
It seems to me that what you really want is an ImageRec constructor that takes
an ImageBuf*. Let me prepare a patch for you that will make this easier. Stay
tuned.
On Mar 27, 2012, at 3:22 AM, Stefan Stavrev wrote:
> I create an ImageBuf and then try to save the image to a file:
>
> ImageBuf* histogram_R = ip_histogram_to_image (r, xres, yres);
> histogram_R->save (path_R);
>
> The result is an image, which is not what it should be. It is
> transparent, and it has some pixels here and there.
>
> I call my function from command line with argument -o output_filename
> so what is left on stack will be saved. To check if histogram_R is ok
> I did the following, which gave the proper result.
>
> ImageSpec specR (256, 256, 1, TypeDesc::FLOAT);
> ot.push (new ImageRec ("irec", specR, ot.imagecache));
> ImageBuf &Rib ((*ot.curimg)());
> Rib = *histogram_R;
>
> Probably I am using "save" the wrong way?
>
> Thanks
--
Larry Gritz
[email protected]
_______________________________________________
Oiio-dev mailing list
[email protected]
http://lists.openimageio.org/listinfo.cgi/oiio-dev-openimageio.org
