[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Message from Mark re: C++




Hi Ilya ,

> but discussing general
> image processing algorithms is of great importance. I feel like I need to
> learn a lot.

So do I, I'm holding the record in things one have to learn :-).

> First, I have to say some words about FM/FiFo vs C++. As far as  know, C is
> rather expensive in case you'd like to be able to redistribute runtimes etc.

Not really, plug ins are plain vanilla dlls in WIN and anything similar 
for MAC AFAIK.
So every C compiler, that is able to compiles a 32 bit dlls and 
to link Windows resources is all you need and all you can use.
There is a free GNU Compiler, Borlands C++ 5.0 should be quite cheep,
and if you are lucky, you may find an old MSVC 4.xx second hand for some
100$.

In my opinion the problem is rather the time, or the manpower
you have to invest to get a certain result (the filter).
For professional work you have to think economical, 
if you can do 80% of all filters with FIFO/FM in 1/4 the time, 
you'll do it that way.

> for me, currently I cannot afford all my time to learning complex C and SDK;
> but I think I can start with FM and maybe this will lead to something
> interesting. That's my opinion.

I'm with you in this point, my opinion is, if one is not particularly 
interested in C programming, there is no reason to try it,
(except you are looking for a new, exciting hobby)
cause it would take a while to find a filter that you can not 
realize in FIFO/FM.
And with just a humbling SDK sample and good ol' Petzold it 
will take you probably 10 month before you see your first filter.

> thanks for the link!.. I think it may be of great use. 

Here are two other sites with links to CG/Image processing stuff, 
the second is quite slow.

http://www.cs.ubc.ca/spider/ladic/imgproc.html 

http://mambo.ucsc.edu/psl/cg.html 

> Interesting kernel; I think asymmetric kernels should always lead to some
> emboss effect, right?

And if you invert the image you'll get the edges.

> calculate the difference between blurred and source image. This difference
> is "edge" picture

Great idea! And the filter looks good.

> difference). My filter also run a "contrast autostretch" 

what do you mean by contrast autostretch? 


regards Thomas