recognize material from sound !

How ? correlation FFT ?
thanks
Is there a theory about ?

On Mar 26, 2:42&#4294967295;pm, "nino" <npkpe...@libero.it> wrote:
> How ? correlation FFT ?
> thanks
> Is there a theory about ?

This can be real hard (impossible) or real easy. Do you know
beforehand how many different materials you need to be able to
identify? And do these "sound" very different from each other or are
they alike?

Back in the days where I worked with color print paper, you would
figure out which side of the paper had the emulsion by rubbing your
finger across it and the emulsion side was quieter than the other
side. But in this case you were able to compare the sound of two
materials against each other.

Clay

On 26 Mar, 19:42, "nino" <npkpe...@libero.it> wrote:
> How ? correlation FFT ?
> thanks
> Is there a theory about ?

Be *very* careful. You can *characterize* the sound
in various ways (decay times, resonance frequencies, etc)
but in general, you can't use that info to *classify*
the sound, i.e. deduce what material or effect caused
the characteristic. The world is an ambiguous place,
so there might be other causes for any given effect
that those you happen to think of.

However, if you have a (small) number of candidates,
you can likely get a long way towards deciding which
of the candidates caused the characteristic. But again,
if whatever characteristic was actually caused by
something else than the alternatives you are looking
for, you have trouble.

Rune

On Mar 26, 2:42&#4294967295;pm, "nino" <npkpe...@libero.it> wrote:
> How ? correlation FFT ?
> thanks
> Is there a theory about ?

There are some forms of this, but you have not stated enough detail to
understand what you want to do.

When I worked in Chemistry a professor was working on Photo Acoustic
Spectroscopy.  A monochromatic light source is shown on the sample in
pulses.  The light is absorbed by the sample in relation to its
wavelength and the properties of the material.  The absorbed light
produces vibrations in the sample which can be detected as sound.  The
sound is analyzed as a function of frequency of the light giving a
spectrograph.  I expect this is not the sort of thing you are looking
for, but you have not given enough info to know what you are thinking
about.

Rick

On Mar 26, 2:42&#4294967295;pm, "nino" <npkpe...@libero.it> wrote:
> How ? correlation FFT ?
> thanks
> Is there a theory about ?

Acoustic signaling has been used to estimate properties
of the earth for geophysical applications in oilfield, etc.
Is this the kind of application that you are asking about?

Julius

Thanks.!
I only want recognize material froma set of 4-5 possibilities by hitting it 
and analizing sound
Thanks
Nini
"rickman" <gnuarm@gmail.com> ha scritto nel messaggio 
news:1b614c17-b411-4051-95b9-7dd71fc742fd@l38g2000vba.googlegroups.com...
On Mar 26, 2:42 pm, "nino" <npkpe...@libero.it> wrote:
> How ? correlation FFT ?
> thanks
> Is there a theory about ?

There are some forms of this, but you have not stated enough detail to
understand what you want to do.

When I worked in Chemistry a professor was working on Photo Acoustic
Spectroscopy.  A monochromatic light source is shown on the sample in
pulses.  The light is absorbed by the sample in relation to its
wavelength and the properties of the material.  The absorbed light
produces vibrations in the sample which can be detected as sound.  The
sound is analyzed as a function of frequency of the light giving a
spectrograph.  I expect this is not the sort of thing you are looking
for, but you have not given enough info to know what you are thinking
about.

Rick 

On 27 Mar, 19:47, "nino" <npkpe...@libero.it> wrote:
> Thanks.!
> I only want recognize material froma set of 4-5 possibilities by hitting it
> and analizing sound

You need to standardize the shape of the objects, as
well as doing both the calibrations and tests in
controlled environments.

The shape will dominate the signatures on a similar
scales as the materials. Imagine comparing the ringing
of two wine glasses, one made of 'plain' glass and
another made of crystal glass. The compare two
glasses of the same kind, but where one has a fracture.

Reflections from walls in the room where the test is
done, and reverberations in small rooms, will also
influence the signatures.

Rune

On Mar 26, 11:42 am, "nino" <npkpe...@libero.it> wrote:
> How ? correlation FFT ?
> thanks
> Is there a theory about ?

Google:

 ultrasonic materials analysis


Dale B. Dalrymple

On Thu, 26 Mar 2009 18:42:44 GMT, "nino" <npkpesae@libero.it> wrote:

>How ? correlation FFT ?
>thanks
>Is there a theory about ?
>

google pitch to midi