Prev: How to put on GUI a web camera
Next: GCE histogram modification algorithm
From: ImageAnalyst on 22 Feb 2010 12:22
Well from what you say, the scaling factors are
a=exp(27) and b=exp(199)
but we haven't read the paper and have no idea what it says. That's
for you to do. Perhaps you can contact the authors for help.

From: Matt J on 24 Feb 2010 09:46
"Parag S. Chandakkar" <parag2489(a)gmail.com> wrote in message <af20b3ff-6858-4a56-9b98-cb18a8fc8fd2(a)w27g2000pre.googlegroups.com>...
> On Feb 21, 9:29 pm, "Matt J " <mattjacREM...(a)THISieee.spam> wrote:

> > Possibly, it's because the maximum value is susceptible to sampling error,or to any noise you have in the images. It would be better to divide the integrals (as approximated by the discrete sums) of both images. This is equivalent to the FFT ratio strategy that you are already using, but on the condition that the max absolute value is attained at DC. I don't know if you verified that.
>
> ------------------------------------------------------------------------------------------------
>
> To reduce the noise and the other errors, I read that you should do
> windowing of images. So I am using Kaiser window. I will test my
> program after that.
>
> I didn't completely understand what you said. Can you explain me in
> detail.
========================

From the formula you gave us

F2(x,y)=1/(a*b) * (F1(x/a,y/b))

Evaluating this at x=y=0 (i.e., at DC) you get

a*b = F1(0,0)/F2(0,0) = Integral(f1)/Integral(f2)

The ratio of the integrals can be approximated by the discrete sums leading to

a*b=sum(f1)/sum(f2)

But the ratio sum(f1)/sum(f2) is much more noise resistant than max(f1)/max(f2)
First  |  Prev  | 
Pages: 1 2
Prev: How to put on GUI a web camera
Next: GCE histogram modification algorithm