Film Works but can Digital do this?

"Roger N. Clark (change username to rnclark)" wrote in
message ...
~~NoMad~~ wrote:
"CharleiD" wrote in message
...
On Wed, 4 Jul 2007 05:49:22 -0600, "~~NoMad~~"

wrote:

Check this amazing photography:
http://www.wired.com/gadgets/digital...5-07/st_nascar

Can this be done with a digital camera?

NM


Nice to see he stole a half-century old method and is trying to act as
if he
invented it. This was used over 50 years ago in 3D aerial photography
for
surveillance of enemy activities during war-times. Recently documented
on a PBS
special. They would fly at speeds of up to 250 mph while photographing
high-resolution images of the land moving beneath the planes that had to
fly
close to the ground to get beneath radar detection. In synchronized 3D
frames no
less. This idiot can't even do that. What a shame that he's doing by
hand what
was done automatically over 50 years ago. Someone should tell this
simpleton and
oaf that there are better ways.

And yes, digital can do even more than this today.


Can you show me a comparable photo to what he as done?

This is the way most spacecraft cameras work, for example,
Landsat Thematic Mapper (TM), and Hyperion cameras
orbiting earth, as well as the EOS spacecraft with numerous
cameras like MODIS; the Mars Reconnaissance Orbiter HIRISE
and CRISM instruments; the Mars Global Surveyor MOC (Mars
orbiter Camera), the Cassini VIMS, CIRS and UVIS cameras
orbiting Saturn; and many others. There are also
aircraft instruments that get images this way too,
e.g. the NASA AVIRIS system. Any of the beautiful images
of earth from earth orbit (most common is Landsat TM) you've
seen are with this line scan technology.

Cameras such as these are basically a line scan camera, and
one uses time to get the second spatial dimension.
In the case of aircraft, it is the plane's forward velocity
that provides the 2nd spatial dimension; for spacecraft,
it is the orbital motion. And all these systems are digital.

Examples:
http://aviris.jpl.nasa.gov/html/aviris.overview.html

The above application of race cars would be much simpler with
a digital line scan camera than with film. In fact, one could
probably rip the guts out of a $200 flatbed scanner and do a better
job and not have to rely on "luck" like with the film setup.

By the way, such line scan digital systems have been in operation
for over 40 years. These are called push-broom scanners.
Think of the bristles of the broom as pixels and you push it
along gathering the second spatial dimension. There are also
"whisk-broom" scanners: one spatial pixel that moves side to side
quickly as you move forward, so you scan both dimensions with
a single pixel to build up an image. The AVIRIS instrument above
is a whisk-broom scanner; TM, MODIS, HIRISE and others mentioned
above are push-broom scanners, and VIMS is a combination of both.
Like pretty much all technology, there are advantages and
disadvantages to each design.

This is pretty cool: this is a replay of the latest Landsat satellite
pass:
http://earthnow.usgs.gov/earthnow_ap...07aa59bff57609
I think it is a real-time playback (clouds usually aren't that
interesting), and the playback is low resolution compared to the
full dataset. You need java and a fast connection.

Roger,

I'm fully aware of line scan photography in aircraft and spacecraft. It
would seem to me that someone would have tried using a line scanner on the
ground something like this guy has done with film. I'll do a search and see
if I can find any examples.

This guy's web site is pretty cool too: http://distavision.com/ I think I
might buy one of his photos.. very unusual.

NM

"~~NoMad~~" wrote in message
...
Check this amazing photography:
http://www.wired.com/gadgets/digital...5-07/st_nascar

Can this be done with a digital camera?

NM


"film zips past the slit at up to 1,400 rpm"

Huh?

So, a little like an old Alpa Roto, but holding the camera straight and not
allowing it to rotate.

Deep.

"Joseph Meehan" wrote in message
...
~~NoMad~~ wrote:
Check this amazing photography:
http://www.wired.com/gadgets/digital...5-07/st_nascar

Can this be done with a digital camera?

NM

Sure. It has been done and is being done. At least one method employs
a scanning system and it is used for air and space survey work.

As for the film system,. my first job was working with a guy who got
his start in photography during WWII processing those long film strips
from the recon planes. We both worked for the same photo studio.


Can you find any examples of a line scan camera used on the ground like this
guy has done?

His website: http://distavision.com/ gives more examples.

Show me something similar done with a digital camera.

NM

In article ,
"~~NoMad~~" wrote:

(snip)


OK, show me a comparable photo.

NM

I just did a google image search for "photo finish" and turned this up:
http://www.goblueraiders.com/backgrounds/trk-finish05-1024.jpg
It looks like a pretty standard output from a digital photo-finish
system.

I assume the ones used for horse and auto racing are faster and more
sophisticated than the ones used for human athletic events, but I assume
they work similarly.


I don't see any reason to crap on the accomplishment described in the
OP's article -- personally I think it's pretty neat for a homebrew
device, and aside from the usual breathless Wired-magazine journalism, I
didn't hear the guy claim that he was the first person to do it or
anything. In fact it seems pretty similar to what Muybridge did back in
the late 19th century, except IIRC he used a moving shutter and a static
film plane; the guy in the Wired article has a moving film plane and
static shutter.

Thinking about the first place I've seen this done *without film*, the
earliest thing that comes to mind are early analog fax machines. Western
Union had a whole system set up for sending faxes that used
photomultiplier tubes (PMTs) and an early drum scanner design.

The sheet you wanted to send was clipped to a small drum, which was
quickly spun. The PMT "reader" was on a clockwork mechanism that scanned
down the sheet, moving slightly over after each rotation of the drum. In
this way, the image on the paper was converted to a changing electrical
signal, which you could send down the wire to a receiving unit, which
had a similar apparatus, except the PMT was replaced with a hot wire
that "burned" the image onto a sheet of thin paper. It's not a line
scanner (it's more of a point scanner) but it's the same principle.

-Kadin.

~~NoMad~~ wrote:
Check this amazing photography:
http://www.wired.com/gadgets/digital...5-07/st_nascar

Can this be done with a digital camera?

Maybe I'm wrong on this but why can't a series of precisely timed,
overlapping photographs, stitched together, deliver the same effect or
something close to this?

~~NoMad~~ wrote:
"Joseph Meehan" wrote in message
...
~~NoMad~~ wrote:
Check this amazing photography:
http://www.wired.com/gadgets/digital...5-07/st_nascar

Can this be done with a digital camera?

NM
Sure. It has been done and is being done. At least one method employs
a scanning system and it is used for air and space survey work.

As for the film system,. my first job was working with a guy who got
his start in photography during WWII processing those long film strips
from the recon planes. We both worked for the same photo studio.


Can you find any examples of a line scan camera used on the ground like this
guy has done?

Simple, try google. There are many applications

http://www.panoscan.com

http://www.rit.edu/~andpph/text-demo-scanner-cam.html

http://adsabs.harvard.edu/abs/1995SPIE.2416..134L

http://jwz.livejournal.com/591617.html

http://news.thomasnet.com/fullstory/485104

http://optoelectronics.perkinelmer.c...=LD3543PGK-022

e.g. google: "line scan camera" digital photography
and see many more. Most applications seem to be industrial.

Roger

"Michael Johnson" wrote in message
...
~~NoMad~~ wrote:
Check this amazing photography:
http://www.wired.com/gadgets/digital...5-07/st_nascar

Can this be done with a digital camera?

Maybe I'm wrong on this but why can't a series of precisely timed,
overlapping photographs, stitched together, deliver the same effect or
something close to this?

Well sort of but then you would have the same still background repeated all
along the strip. It may be possible to digitally remove the still background
image.

Hmmm...

NM

"~~NoMad~~" wrote in message
...
Check this amazing photography:
http://www.wired.com/gadgets/digital...5-07/st_nascar

Can this be done with a digital camera?

NM



OK, with a bit of research I figure a digital camera or line scanner can't
reproduce this image because the recording speed is too slow. With a digital
line scanner you would need to record all pixels in parallel at the same
instant to prevent distortion. Now even if you did have a parallel recording
line scanner you would have to record to memory at about 1 gigabyte per
second for reasonable resolution. Don't forget the color too. This speed
requirement is just too fast for current technology.

So for this application Film Still Rules!

NM

~~NoMad~~ wrote:
"Michael Johnson" wrote in message
...
~~NoMad~~ wrote:
Check this amazing photography:
http://www.wired.com/gadgets/digital...5-07/st_nascar

Can this be done with a digital camera?
Maybe I'm wrong on this but why can't a series of precisely timed,
overlapping photographs, stitched together, deliver the same effect or
something close to this?

Well sort of but then you would have the same still background repeated all
along the strip. It may be possible to digitally remove the still background
image.

Hmmm...

Isn't his camera stationary and the film is moving in a continuous
motion? It looks like the background is the same and he is essentially
stitching together narrow slices of images. What changes is the
subjects moving through the narrow slit while the background is the same
in every image slice. Now I realize that using film is similar to an
analog recording verses a time slice for digital recording but the
concept of capturing the image is the same. Doing the same thing with a
digital camera is a matter of timing the shutter speed verses the films
linear speed moving past the slit. I would think it is easier to
accomplish this digitally.

"Roger N. Clark (change username to rnclark)" wrote in
message ...
~~NoMad~~ wrote:
"Joseph Meehan" wrote in message
...
~~NoMad~~ wrote:
Check this amazing photography:
http://www.wired.com/gadgets/digital...5-07/st_nascar

Can this be done with a digital camera?

NM
Sure. It has been done and is being done. At least one method
employs a scanning system and it is used for air and space survey work.

As for the film system,. my first job was working with a guy who got
his start in photography during WWII processing those long film strips
from the recon planes. We both worked for the same photo studio.


Can you find any examples of a line scan camera used on the ground like
this guy has done?

Simple, try google. There are many applications

http://www.panoscan.com

http://www.rit.edu/~andpph/text-demo-scanner-cam.html

http://adsabs.harvard.edu/abs/1995SPIE.2416..134L

http://jwz.livejournal.com/591617.html

http://news.thomasnet.com/fullstory/485104

http://optoelectronics.perkinelmer.c...=LD3543PGK-022

e.g. google: "line scan camera" digital photography
and see many more. Most applications seem to be industrial.

Roger,

Unfortunately none of these line scan cameras are fast enough to capture the
NASCAR racers....

NM