White House roof edited in USGS photos

In a previous article, Mxsmanic said:
Steve Andrew writes:
Really ? - I suggest you do your math and play around with the Rayliegh
criteria. Try calculating the diameter of the lens required to resolve a
wris****ch for an altitude of around 250-300km. It is generally accepted by
those who know that current resolution is in the order of 10cm.

No, it is generally _acknowledged_ that the current resolution is in
this range. Actual resolutions tend to be classified. Spy satellites
long ago abandoned purely optical imaging.

If there is one thing this thread, and the ones on rec.aviation.piloting
about the SR-71 have taught me, it's that the vast majority of Usenet
posters believe that the US intelligence community and military are not
bound by the laws of physics.

--
Paul Tomblin http://xcski.com/blogs/pt/
The problem with eating a high-fibre diet is that, before you know it, your
lower bowel will have attracted a whole mass of backhoes.
-- Tanuki

Dave Head writes:

Would, if it kills the pilot, or disrupts the electronics or hydraulics, which
seems likely considering the hail of bullets.

No, it wouldn't. The 767 would hit the ground in pieces and flames
either way.

--
Transpose hotmail and mxsmanic in my e-mail address to reach me directly.

Steve Andrew writes:

Hmmmm... interesting. If not optical, then what alternatives are there,
apart from radar?.

Digital image processing.

Can you provide any links that might expand on this ?

No.

--
Transpose hotmail and mxsmanic in my e-mail address to reach me directly.

Paul Tomblin writes:

If there is one thing this thread, and the ones on rec.aviation.piloting
about the SR-71 have taught me, it's that the vast majority of Usenet
posters believe that the US intelligence community and military are not
bound by the laws of physics.

Who said anything about the US?

--
Transpose hotmail and mxsmanic in my e-mail address to reach me directly.

Paul Cooper wrote:
On Wed, 12 May 2004 03:37:35 GMT, "Steve Andrew"
wrote:


Mxsmanic wrote:

Steve Andrew writes:


Really ? - I suggest you do your math and play around with the
Rayliegh criteria. Try calculating the diameter of the lens required
to resolve a wris****ch for an altitude of around 250-300km. It is
generally accepted by those who know that current resolution is in
the order of 10cm.

No, it is generally _acknowledged_ that the current resolution is in
this range. Actual resolutions tend to be classified. Spy satellites
long ago abandoned purely optical imaging.

Hmmmm... interesting. If not optical, then what alternatives are there,
apart from radar?.

As an electronics engineer I'm confident in saying that even milli-metric
radar will not provide the resolution, nor the signal return required to
read the time on a wrist-watch from 100-200kM.

Can you provide any links that might expand on this ?



As radars are all longer wavelengths than optical, the Rayleigh
criterion gets you even harder there. And although larger apertures
can be synthesized with radar, it is no longer an instantaneous view,
so movement cooks your goose. The Rayleigh criterion rules, and the
stated best resolution of 10 cm is the best physically available. You
might beat it with optical interferometry, but any motion on the
ground will wreck your image. Given that the atmosphere is not that
stable, I'd say you were at the bleeding edge at 10cm with a satellite
mountable optical system.

Incidentally, the other restriction is in downlink band-width. A
softer limit than the hard physics gives you, but still a limit to be
considered. At 10cm pixel size, a 100m square is a megapixel!

It all depends if you're talking about real-time or near-realtime.

Think about it. At 10cm resolution, capture a lot of images. They're not
going to be taken from the exact same place and won't be entirely
identical. The edges of the pixels won't perfectly overlap between the
images, and from the differences in the overlap, you can extract a lot
of data not visible in the individual captures. Of course, the more you
enhance it, the more abstracted you will get, past a certain point, and
artifacts of the processing will start to generate at a certain point.

Of course, since these images are captured over time, it might be
possible to determine the type of a watch, but actually reading the time
won't be possible, since it won't be the same time in any of the
sequence of captures. ;-)





--
The incapacity of a weak and distracted government may
often assume the appearance, and produce the effects,
of a treasonable correspondence with the public enemy.
--Gibbon, "Decline and Fall of the Roman Empire"

In rec.photo.digital Steve Andrew wrote:
Paul Cooper wrote:
On Wed, 12 May 2004 03:37:35 GMT, "Steve Andrew"
wrote:

Mxsmanic wrote:
Steve Andrew writes:

Really ? - I suggest you do your math and play around with the
Rayliegh criteria. Try calculating the diameter of the lens
required to resolve a wris****ch for an altitude of around
250-300km. It is generally accepted by those who know that current
resolution is in the order of 10cm.

No, it is generally _acknowledged_ that the current resolution is in
this range. Actual resolutions tend to be classified. Spy
satellites long ago abandoned purely optical imaging.

Hmmmm... interesting. If not optical, then what alternatives are
there, apart from radar?.

As an electronics engineer I'm confident in saying that even
milli-metric radar will not provide the resolution, nor the signal
return required to read the time on a wrist-watch from 100-200kM.

Can you provide any links that might expand on this ?


As radars are all longer wavelengths than optical, the Rayleigh
criterion gets you even harder there. And although larger apertures
can be synthesized with radar, it is no longer an instantaneous view,
so movement cooks your goose. The Rayleigh criterion rules, and the
stated best resolution of 10 cm is the best physically available. You
might beat it with optical interferometry, but any motion on the
ground will wreck your image. Given that the atmosphere is not that
stable, I'd say you were at the bleeding edge at 10cm with a satellite
mountable optical system.

Incidentally, the other restriction is in downlink band-width. A
softer limit than the hard physics gives you, but still a limit to be
considered. At 10cm pixel size, a 100m square is a megapixel!

Just in case some bright spark asks, no you can't use ultra-violet!
The atmosphere is (for practical purposes) opaque at wavelengths
shorter than the optical band.

Paul

Thanks for that Paul. As an engineer I've learned over the years to never
fall into the trap of saying something can never be done. Having said that,
some things *cannot* be done, one of them being changing the laws of physics


Now let's wait for somebody to mention quantum mechanics... ;-

Steve

Won't help. It adds a bit of uncertainty... ;-)

---- Paul J. Gans

Mxsmanic wrote:

The White House is quite a source of bad ideas these days, though.


Jeez. Ain't it the truth.

"Mxsmanic" wrote in message
...

There are plenty of spy satellites in the world, some of why can read
the time on a person's watch.

Didn't you just love that "Enemy of the State" movie?

Apart from all the physical limitations mentioned in other posts, it is
interesting to note the big bonanza that the commercial purveyors of high
resolution satellite images are in right now. IKONOS, QuickBird and the like
are seeing volumes of sales not seen before. This increase in sales is
largely due to large contracts awarded to the different American companies
by NIMA, which is part of the DOD. In fact, the contracts are so large (in
the order of $250M a piece) that these commercial companies can leverage
their size to build the next generation of optical satellites.

Now why would the cash-strapped DOD spend up to $1B in commercial imagery if
they have all the spy satellites with far greater accuracy? Part of the
answer is in the fact that the current spy satellites are all old and
running out of fuel (which is necessary to support orbital changes to direct
coverage to hotspots). Another part of the answer is that the capabilities
of the spy satellites are simply not that much higher than the commercial
optical systems. (The last part would be the support of a commercial sector
to offload the burden from the military to the private sector. This policy
dates back to the Clinton administration and is well documented for Landsat
follow-up, for instance, but it applies equally to other optical systems.)

Patrick

On Wed, 12 May 2004 09:55:57 -0400, Tiny Human Ferret
wrote:

Paul Cooper wrote:


As radars are all longer wavelengths than optical, the Rayleigh
criterion gets you even harder there. And although larger apertures
can be synthesized with radar, it is no longer an instantaneous view,
so movement cooks your goose. The Rayleigh criterion rules, and the
stated best resolution of 10 cm is the best physically available. You
might beat it with optical interferometry, but any motion on the
ground will wreck your image. Given that the atmosphere is not that
stable, I'd say you were at the bleeding edge at 10cm with a satellite
mountable optical system.

Incidentally, the other restriction is in downlink band-width. A
softer limit than the hard physics gives you, but still a limit to be
considered. At 10cm pixel size, a 100m square is a megapixel!

It all depends if you're talking about real-time or near-realtime.

Think about it. At 10cm resolution, capture a lot of images. They're not
going to be taken from the exact same place and won't be entirely
identical. The edges of the pixels won't perfectly overlap between the
images, and from the differences in the overlap, you can extract a lot
of data not visible in the individual captures. Of course, the more you
enhance it, the more abstracted you will get, past a certain point, and
artifacts of the processing will start to generate at a certain point.

Of course, since these images are captured over time, it might be
possible to determine the type of a watch, but actually reading the time
won't be possible, since it won't be the same time in any of the
sequence of captures. ;-)

A very nice idea that has been used to enhance the resolution of
images taken on Mars. However it won't work in this case for the
following reasons:

1) The atmosphere isn't stable enough to allow detailed registration
of the images
2) Most targets you'd be interested in seeing that much detail of are
moving! The techniques relies on the images being the same repeatedly,
and in this case, they aren't.

Paul

"Steve Andrew" wrote in message
...
Mxsmanic wrote:
There are plenty of spy satellites in the world, some of why can read
the time on a person's watch.

Really ? - I suggest you do your math and play around with the Rayliegh
criteria. Try calculating the diameter of the lens required to resolve a
wris****ch for an altitude of around 250-300km. It is generally accepted
by
those who know that current resolution is in the order of 10cm.

The Rayleigh effect is so highly dependent on wavelength that it's effect is
sharply dimished in the red and infrared wavelength regions. So for any
object that can be detected without having to use blue light reflection the
Rayleigh effect is not the overriding constraint.

The atmospheric turbidity has a much more pronounced effect for any
practical application IMHO. Stars don't twinkle, the atmosphere just blurrs
their beam. This is BTW also something that is being overcome, although not
yet at the satellite scale AFAIK. The Hubble space telescope is no longer
maintained because the latest generation of ground-based actively corrected
optical telescopes (several meters diameter mirror) are approaching or even
exceeding the clarity of Hubble. Older techniques were more compact and use
a laser to establish the real-time atmospheric properties and correct for
distortion in software. Such an approach would be very viable for a
satellite, since lasers are compact, solid state, and low maintenance, while
all the correction can be done in the terrestrial segment. I would be very
suprised if such correction is not applied to imagery from spy satellites. I
still don't think that the resolution will be sufficient to zoom in on all
the physiological details of the proverbial suntanning topless babe (e.g.
"On deadly ground II" movie, which has a few more physical impossibilities
attributes to orbital vehicles), but one day we just may get there.

Telling the time from space? Just buy a watch!

Patrick