| Topic: |
Science > Physics |
| User: |
"" |
| Date: |
27 Mar 2007 08:42:10 PM |
| Object: |
Transform optics |
I'm learning something related to transform optics, but I don't even
have any basic ideas about that. Would you please give me some hints
if anyone konws?
Thanks in advance!
.
|
|
| User: "Androcles" |
|
| Title: Re: Transform optics |
28 Mar 2007 03:23:26 PM |
|
|
<Lefthanded.Material@gmail.com> wrote in message =
news:1175046130.816660.207520@n59g2000hsh.googlegroups.com...
I'm learning something related to transform optics, but I don't even
have any basic ideas about that. Would you please give me some hints
if anyone konws?
=20
Thanks in advance!
"Transform" might be a good place to start. That is a mathematical term.
If you've ever heard of a pin-hole camera then you'd know that the image =
is inverted.
http://en.wikipedia.org/wiki/Pinhole_camera
The transform in this case is a mathematical function which describes
how a 40 ft high three-dimensional tree appears as a 1 inch high two-
dimensional inverted image on a photograph.
The scale is quite simple, it is the distance from the pinhole
to the back of the camera divided by the distance from the pinhole
to the tree. Thus "transform optics" is the mathematics of optics.
It gets a little more complicated than that when we use lenses
and holograms, but it is still mathematics.
http://en.wikipedia.org/wiki/Transformation_(mathematics)
.
|
|
|
|
| User: "Tom Potter" |
|
| Title: Re: Transform optics |
29 Mar 2007 07:35:58 AM |
|
|
<Lefthanded.Material@gmail.com> wrote in message news:
1175046128.571921.235660@y80g2000hsf.googlegroups.com...
I'm learning something related to transform optics, but I don't even
have any basic ideas about that. Would you please give me some hints
if anyone konws?
Thanks in advance!
You could find a lot of useful information
with a U.S. Patent search on "optical correlator".
http://patft.uspto.gov/netahtml/PTO/search-bool.html
An "optical correlator" is a system composed of
a monochromatic light source, holograms and lenses.
Unlike digital computers that process a string of data,
"optical correlators" process an "area" of data.
For example they can perform a real time fourier transforn
on an area of data in real time.
They have been used for many years to
process optical and RADAR data.
For example, back in the 1960's
one of the first optical correlators spotted the
missile being set up in Cuba, when analyzing
side-looking RADAR images of Cuba.
As LASERS were not yet invented,
the first optical correlators used filters
to get the monochromatic light needed.
"Signature" holograms can be constructed
to find almost any kind of target,
such as planes, trucks, construction equipment, etc.
and optical correlator systems can be used to
automatically sort out the changes in RADAR data or photos
taken at different times.
Much of this is highly classified,
but there are many commercial patents on optical correlators.
--
Tom Potter
*** Time Magazine Person of the Year 2006 ***
http://home.earthlink.net/~tdp/
http://tdp1001.googlepages.com/home
http://no-turtles.com
http://www.frappr.com/tompotter
http://photos.yahoo.com/tdp1001
http://spaces.msn.com/tdp1001
http://www.flickr.com/photos/tom-potter/
http://tom-potter.blogspot.com
.
|
|
|
|
| User: "Randy Poe" |
|
| Title: Re: Transform optics |
28 Mar 2007 04:06:34 PM |
|
|
On Mar 27, 9:42 pm, wrote:
I'm learning something related to transform optics, but I don't even
have any basic ideas about that. Would you please give me some hints
if anyone konws?
Thanks in advance!
You might find "Fourier optics" a more productive search term. Optical
devices perform Fourier transforms on their illumination patterns.
For instance, the diffraction pattern of a slit is the Fourier
transform
of the illumination of that slit.
- Randy
.
|
|
|
| User: "Androcles" |
|
| Title: Re: Transform optics |
28 Mar 2007 05:35:57 PM |
|
|
"Randy Poe" <poespam-trap@yahoo.com> wrote in message =
news:1175115994.385064.101020@b75g2000hsg.googlegroups.com...
On Mar 27, 9:42 pm, wrote:
I'm learning something related to transform optics, but I don't even
have any basic ideas about that. Would you please give me some hints
if anyone konws?
Thanks in advance!
=20
You might find "Fourier optics" a more productive search term. Optical
devices perform Fourier transforms on their illumination patterns.
For instance, the diffraction pattern of a slit is the Fourier
transform
of the illumination of that slit.
=20
- Randy
Why don't you confuse the kid, fuckhead?
Pretending you are smart for the greater glory of Poe?
.
|
|
|
| User: "ZHANG Pu" |
|
| Title: Re: Transform optics |
28 Mar 2007 09:02:32 PM |
|
|
On Mar 29, 6:35 am, "Androcles" <Engin...@hogwarts.physics.co.uk>
wrote:
"Randy Poe" <poespam-t...@yahoo.com> wrote in messagenews:1175115994.385064.101020@b75g2000hsg.googlegroups.com...
On Mar 27, 9:42 pm, wrote:
I'm learning something related to transform optics, but I don't even
have any basic ideas about that. Would you please give me some hints
if anyone konws?
Thanks in advance!
You might find "Fourier optics" a more productive search term. Optical
devices perform Fourier transforms on their illumination patterns.
For instance, the diffraction pattern of a slit is the Fourier
transform
of the illumination of that slit.
- Randy
Why don't you confuse the kid, fuckhead?
Pretending you are smart for the greater glory of Poe?
Maybe I didn't fomulate my question clearly. But anyway Randy's reply
is
closer to what I want.
Can you tell me whether Fourier Optics deals with co-ordinate
transformation?
Thank you, both guys.
.
|
|
|
| User: "Androcles" |
|
| Title: Re: Transform optics |
29 Mar 2007 05:56:36 AM |
|
|
"ZHANG Pu" <Lefthanded.Material@gmail.com> wrote in message =
news:1175133752.498484.150340@p77g2000hsh.googlegroups.com...
On Mar 29, 6:35 am, "Androcles" <Engin...@hogwarts.physics.co.uk>
wrote:
"Randy Poe" <poespam-t...@yahoo.com> wrote in =
messagenews:1175115994.385064.101020@b75g2000hsg.googlegroups.com...
On Mar 27, 9:42 pm, wrote:
I'm learning something related to transform optics, but I don't =
even
have any basic ideas about that. Would you please give me some =
hints
if anyone konws?
Thanks in advance!
You might find "Fourier optics" a more productive search term. =
Optical
devices perform Fourier transforms on their illumination patterns.
For instance, the diffraction pattern of a slit is the Fourier
transform
of the illumination of that slit.
- Randy
Why don't you confuse the kid, fuckhead?
Pretending you are smart for the greater glory of Poe?
=20
Maybe I didn't fomulate my question clearly.=20
Yes you did, it was very clear. Maybe not what you intended,
but I gave you a clear reply to what you clearly asked.=20
But anyway Randy's reply
is
closer to what I want.
Can you tell me whether Fourier Optics deals with co-ordinate
transformation?
When we add two or more frequencies together we get a "signal"
which is the sum of the frequencies.
A Fourier transformation does the opposite, it extracts the pure=20
sinusoidal frequencies from the signal so that we understand=20
what the signal is composed of. Frequencies do not have cartesian=20
coordinates but sometimes we simulate time with distance so that=20
it appears they do.
For example, this decaying sine wave:=20
http://www.kettering.edu/~drussell/Demos/SHO/damp.html
is shown using cartesian coordinates but the mass and spring
do not move in x at all, they only move in y and time.=20
In this simulation
http://www.kettering.edu/~drussell/Demos/BeerBottle/beerbottle.html
the bottle goes nowhere but the chart shows frequency plotted=20
against sound pressure.=20
Optically, white light is split into its component frequencies=20
by a prism.=20
http://school.discovery.com/clipart/images/prism4c.gif
and so a prism does a Fourier transformation of the "signal".
It does NOT perform a coordinate transformation, whereas
a pinhole camera does. They are completely different types
of optical transformations, I do not have clairvoyance to=20
know which type you are looking for.
As I stated, the first think you need to understand is what
a transformation is. Poe has heard of Fourier transformations
and he therefore assumes that is what you want. He's a
*****, there are many.=20
.
|
|
|
|
| User: "Randy Poe" |
|
| Title: Re: Transform optics |
29 Mar 2007 08:01:04 AM |
|
|
On Mar 28, 10:02 pm, "ZHANG Pu" <> wrote:
On Mar 29, 6:35 am, "Androcles" <Engin...@hogwarts.physics.co.uk>
wrote:
"Randy Poe" <poespam-t...@yahoo.com> wrote in messagenews:1175115994.385064.101020@b75g2000hsg.googlegroups.com...
On Mar 27, 9:42 pm, wrote:
I'm learning something related to transform optics, but I don't even
have any basic ideas about that. Would you please give me some hints
if anyone konws?
Thanks in advance!
You might find "Fourier optics" a more productive search term. Optical
devices perform Fourier transforms on their illumination patterns.
For instance, the diffraction pattern of a slit is the Fourier
transform
of the illumination of that slit.
- Randy
Why don't you confuse the kid, fuckhead?
Pretending you are smart for the greater glory of Poe?
Maybe I didn't fomulate my question clearly. But anyway Randy's reply
is
closer to what I want.
Can you tell me whether Fourier Optics deals with co-ordinate
transformation?
Not that I've ever heard of, but that doesn't necessarily rule it out.
What sort of coordinate transformation?
Why don't you start with reading some tutorials on Fourier
optics and see if that's what you had in mind?
Optical computing was a hot topic of research about 15
or 20 years ago because of the ability of optical systems
to do certain operations (like FTs) quickly. People looked
at ways to do all sorts of computations optically. When I
do a Google search for "optical computing coordinate
transformation" I turn up several references that might be
of interest to you.
- Randy
.
|
|
|
|
| User: "Andy Resnick" |
|
| Title: Re: Transform optics |
29 Mar 2007 08:58:52 AM |
|
|
ZHANG Pu wrote:
<snip>
Maybe I didn't fomulate my question clearly. But anyway Randy's reply
is
closer to what I want.
Can you tell me whether Fourier Optics deals with co-ordinate
transformation?
Thank you, both guys.
Fourier optics is a way to describe optical (spatial) signal processing,
making tha analogy to time-varying signal processing. That is, rather
than describe the object as a collection of luminous points, it is
equivalently described in terms of periodic functions. That is, we can
either have a function g(x,y) to describe the intensity at each point
(x,y), or equivalently have the function G(u,v) which is the Fourier
transform of g(x,y). To be sure, actual optical systems are slightly
more complex, but the concept is simple enough- Laue patterns, Kossel
lines, and Bragg scattering are all frequency-domain methods.
The amazing thing, IMO, is that a simple lens can perform a 2D Fourier
transform of the optical field. The image can then be operated on in
Fourier space, emphasizing or de-emphasizing particular spatial
frequencies, to for example, enhance edges or reduce noise.
--
Andrew Resnick, Ph.D.
Department of Physiology and Biophysics
Case Western Reserve University
.
|
|
|
|
|
|
|
|
Related Articles |
|
|
