| Re: Image fusion algorithm... [message #32661 is a reply to message #32569] |
Wed, 23 October 2002 09:33  |
Karl Schultz
Messages: 341
Registered: October 1999
|
Senior Member |
|
|
"Gonzalo Rojas" <rojas_gonzalo_news@hotmail.com> wrote in message
news:3db5f86e$1_2@nova.entelchile.net...
> "Dick Jackson" <dick@d-jackson.com> wrote in
> news:1cWs9.546844$v53.23181338@news3.calgary.shaw.ca:
>
>> "Gonzalo Rojas" <rojas_gonzalo_news@hotmail.com> wrote in message
>> news:3db40db3$1_2@nova.entelchile.net...
>>> Hi:
>>>
>>> Do you know what algorithm uses the image blend functionality of IDL
>>> 5.5?...
>>
>> If you're asking about the Object Graphics class IDLgrImage and its
>> blending function options, you'll find it in Online Help under
>> "IDLgrImage:class Init method", "Blend_Function" keyword. There are
>> several algorithms available.
>>
>> For a coherent explanation of this issue and much else, see Ronn Kling's
>> excellent new book "Power Graphics with IDL".
>> http://www.kilvarock.com/books/powergraphics.htm
>>
>> Cheers,
>> --
>> -Dick
>>
>> Dick Jackson / dick@d-jackson.com
>> D-Jackson Software Consulting / http://www.d-jackson.com
>> Calgary, Alberta, Canada / +1-403-242-7398 / Fax: 241-7392
>>
>>
>
> Hi Dick:
>
> Yes... I am asking about the blending function of Object Graphics class
> IDLgrImage...
>
> I read the online help, but I need to know details such as: how the
> algorithm does the blending between two true color images ?... Could the
> algorithm do such blending ?...
It really doesn't blend between two images. The image in question (the
source) is blended with what is already on the screen (the destination).
So, if you wanted to blend two images, you'd draw one to the screen with no
blending and then draw the other image with your blend factor (alpha).
Is your question about color components? Each component is blended
independently according to the discussion in the docs for BLEND_FUNCTION.
For example, a common choice for the blend function is (3,4) which produces
traditional "alpha blending". This gives:
C(dest)' = (ImageAlpha * C(image)) + ((1-ImageAlpha) * C(dest))
This equation is applied to *each* of the R, G, and B components
independently. "C" in the above equation means one of the color components.
The docs call "C" a pixel, but it really means each color component or
channel of the pixel.
Note that you can't do blending in indexed mode.
As far as the actual implementation goes, IDL doesn't actually do the
blending. It tells OpenGL to perform the blending in the requested manner
and then IDL tells OpenGL to write the image to the screen. OpenGL then
modifies the pixels on the screen according to the specified blend function,
the input (image) pixels and what's already on the screen.
I still may not understand your question, but I hope that these explanations
help.
Karl
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
