Re: Mean filter

[Quentin Spencer]

Søren Hauberg wrote:

> Robert A. Macy wrote:
>
> > Was that the first time you ran meanfilter? 
> > Doesn't the second time run faster?
>
> No that was the third time.
> In the test below I cheated a little bit and created the filter (f) 
> before I called tic, so this will also affect time.
>
> Thanks,
> Søren
>
> > > Thanks, I hadn't thought of that implementation - it's
> > > nice :-)
> > > Unforunaly it doesn't appear to be fast enough:
> > >
> > > a = rand(1,100000);
> > > f = ones(1,100)/100;
> > >
> > > tic; m1 = meanfilter(a, 100); toc
> > > ans = 0.15448
> > >
> > > tic; m2 = conv(a, f); toc
> > > ans = 0.12615

I tried this with both the filter and conv functions, with similar 
results (though filter seemed very slightly faster). Since they are 
precompiled functions, I think the only way you could improve on them is 
to write your own precompiled function that does only addition and no 
multiplication--you could scale the results afterward if you needed to.

-Quentin



-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:  http://www.octave.org
How to fund new projects:  http://www.octave.org/funding.html
Subscription information:  http://www.octave.org/archive.html
-------------------------------------------------------------