Re: Camera software
Posted by
Elliot Burke
on 2006-08-02 17:37:19 UTC
Dennis,
If you need wideband white noise (white in the spatial frequency sense),
wouldn't laser speckle work for you? A laser beam passing through a piece
of ground glass will project a speckle pattern. Focusing the laser beam to
different sizes on the ground glass will change the size the speckle: small
laser beam->large speckle, large laser beam -> small speckle.
Changing the position of the ground glass slightly ( a fraction of the beam
size) will change the speckle pattern completely.
Speckle measurement is used in several types of optical mouse, search for
"laser mouse".
It isn't actually white noise, more pink, the math is very interesting but
this is not the right place to go into it.
Using speckle for measurement has been done to sub-angstrom levels by those
who really really try hard.
If you are interested in using structured light, you might want to look at
FringeXP (freeware). This is used to measure optical surfaces
interferometrically, but will work some structured light applications.
regards-
Elliot
Re: Camera software
Posted by: "Dennis Schmitz" denschmitz@... denschmitz
Wed Aug 2, 2006 2:08 pm (PST)
I'm considering prototyping one using a couple of webcams mounted on
my Sherline, assuming my math is right and it'll be worth the time.
It'll probably be a sloppy mess like you find in a research lab
though.
I'll stick with stereo cameras because depth map software works better
if the light source is fixed relative to the object that's moving. In
fact, it's better for the camera to move around the object unless
you're using structured light illumination (I think one of the
in-development system uses slit illuminator.)
Stereo software works by triangulating features in the image from two
views, so there isn't need to project anything on the image except to
enhance surface texture, in which case you could use anything, really.
Wideband white noise that changes every frame would be best, but I
haven't found a cheap way to do that.
Right now I'm looking at two CMOS cameras with around 3Mpixels and an
LED light source. And software. That plus a way to calibrate the whole
system, like a set of gauge blocks.
--
Internal Virus Database is out-of-date.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.10.4/399 - Release Date: 7/25/2006
If you need wideband white noise (white in the spatial frequency sense),
wouldn't laser speckle work for you? A laser beam passing through a piece
of ground glass will project a speckle pattern. Focusing the laser beam to
different sizes on the ground glass will change the size the speckle: small
laser beam->large speckle, large laser beam -> small speckle.
Changing the position of the ground glass slightly ( a fraction of the beam
size) will change the speckle pattern completely.
Speckle measurement is used in several types of optical mouse, search for
"laser mouse".
It isn't actually white noise, more pink, the math is very interesting but
this is not the right place to go into it.
Using speckle for measurement has been done to sub-angstrom levels by those
who really really try hard.
If you are interested in using structured light, you might want to look at
FringeXP (freeware). This is used to measure optical surfaces
interferometrically, but will work some structured light applications.
regards-
Elliot
Re: Camera software
Posted by: "Dennis Schmitz" denschmitz@... denschmitz
Wed Aug 2, 2006 2:08 pm (PST)
I'm considering prototyping one using a couple of webcams mounted on
my Sherline, assuming my math is right and it'll be worth the time.
It'll probably be a sloppy mess like you find in a research lab
though.
I'll stick with stereo cameras because depth map software works better
if the light source is fixed relative to the object that's moving. In
fact, it's better for the camera to move around the object unless
you're using structured light illumination (I think one of the
in-development system uses slit illuminator.)
Stereo software works by triangulating features in the image from two
views, so there isn't need to project anything on the image except to
enhance surface texture, in which case you could use anything, really.
Wideband white noise that changes every frame would be best, but I
haven't found a cheap way to do that.
Right now I'm looking at two CMOS cameras with around 3Mpixels and an
LED light source. And software. That plus a way to calibrate the whole
system, like a set of gauge blocks.
--
Internal Virus Database is out-of-date.
Checked by AVG Free Edition.
Version: 7.1.394 / Virus Database: 268.10.4/399 - Release Date: 7/25/2006
Discussion Thread
Dennis Schmitz
2006-07-31 13:35:59 UTC
Camera software
Chris Horne
2006-07-31 14:21:00 UTC
Re: Camera software
ballendo
2006-07-31 23:25:48 UTC
Re: Camera software
George Taylor, IV
2006-08-01 00:12:06 UTC
RE: [CAD_CAM_EDM_DRO] Camera software
th_carel
2006-08-01 01:17:36 UTC
Re: [CAD_CAM_EDM_DRO] Camera software
Elliot Burke
2006-08-01 06:48:02 UTC
Re: Camera software
Dennis Schmitz
2006-08-02 12:17:41 UTC
Re: [CAD_CAM_EDM_DRO] Re: Camera software
wanliker@a...
2006-08-02 13:11:26 UTC
Camera software
Dennis Schmitz
2006-08-02 14:08:27 UTC
Re: Camera software
Elliot Burke
2006-08-02 17:37:19 UTC
Re: Camera software
Graham Stabler
2006-08-03 02:24:58 UTC
Re: Camera software
Dennis Schmitz
2006-08-03 04:05:20 UTC
Re: [CAD_CAM_EDM_DRO] Re: Camera software
Graham Stabler
2006-08-03 05:20:55 UTC
Re: Camera software
Dennis Schmitz
2006-08-03 06:40:25 UTC
Re: Camera software
Graham Stabler
2006-08-03 08:07:04 UTC
Re: Camera software
wthomas@g...
2006-08-03 21:28:46 UTC
W.E.T.Re: [CAD_CAM_EDM_DRO] Edge Finder
Graham Stabler
2006-08-04 03:01:59 UTC
W.E.T.Re: [CAD_CAM_EDM_DRO] Edge Finder