![[Home]](http://www.bayleshanks.com/cartoonbayle.png)
Apparently, six years ago, U.S. spy satellites could resolve objects as small as a baseball (~10 cm).
As of 2001 an expert said that U.S. spy satellites could probably resolve an individual person but not well enough to recognize them, or to read a license plate.
Can spy satellites read license plates today? My bet is yes.
Can spy satellites see through walls? I'd say maybe (although obviously not as well as they can see stuff out in the open).
Commercial satellites can resolve at least half-meter objects.
"commercial spy satellites" can do 10-inch resolution.
I expect secret U.S. govm't satellites can do much better.
Apparently, six years ago, U.S. spy satellites could resolve objects as small as a baseball (~10 cm) (another source: a nytimes article clipping that i have around here somewhere but i can't find right now).
As of 2001 an expert said that U.S. spy satellites could probably resolve an individual person but not well enough to recognize them, or to read a license plate.
Can current spy satellites see license plates? My bet is yes.
Clouds. Apparently there are imaging techniques to see through clouds. I bet you can see underground and through walls if you settle for non-visible light wavelengths. Note that thermal imaging can give you a lot of information about what is being done where visible light can't penetrate example.
Angle of view. I've heard that you don't have to look directly downwards with your spy satellite (so that fact that license plates aren't facing the sky isn't a show-stopper).
Dawes limit on resolution based on objective size. I don't know much about optics, but my impression is that people do beat these "theoretical limits" on objective resolution with tricks like averaging over multiple pictures taken at different times, or having multiple separate lenses whose images are combined{{ Incidentally, just today I was reading a review article in biology that supports a skepticism of theoretical limits: "Finally, novel approaches are challenging even the most basic assumptions of light microscopy, such as the diffraction limit of spatial resolution, always viewed as immutable. Hyper-resolution microscopies, or stimulated-emission depletion strategies, have pushed the spatial resolution of light microscopy to tens of nanometers in special cases. Although these new methods have not yet been applied systematically to living samples, they allow microscopists to achieve milestones that were supposedly unachievable, such as optically monitoring the spatiotemporal dynamics of molecules well below the wavelength of light."}}.
Anyway, you can do weird stuff with optics. In neuroscience, if you have a slice of brain, you can "focus down" a few millmeters PAST THE SURFACE of the top of the slice, and still get sharp resolution. And that's just using run-of-the-mill microscopes. There are special ones that can go deeper. See here and here for some other kinds of counterintuitive "tricks".
It's illegal to spy domestically. I think the US govm't is prohibited by law from spying within US borders (although this has probably been loosened by PATRIOT) but I've heard we routinely trade information with the Brits and others (i.e. we give them our pictures within their borders, and they reciprocate; that way everyone gets all the pictures and no one has to break the anti-surveillance laws).
It's too much information to comb through. I think this is the current limiting factor, just as with the government satellites that monitor cell phone calls and the government taps at internet backbones.
But we're working on it. This is exactly what the "data mining" subfield of A.I. is good for. By the way, the DOD is the major funder of A.I. research, and the last time I checked (2 yrs ago), there was a preponderance of grant solicitations specifically for data-mining in ways that would be useful for surveillance; also projects like automated face recognition (the DOD is also openly interested in developing autonomous robot soldiers, so that's another reason they fund A.I.). And as we know, the intelligence community is actively and semi-openly researching social network analysis methods ("friend of a friend" stuff) including automated reconstruction of social networks from IM, email, phone logs, etc.
So what I'm saying is, the data mining problem is already solved for many applications, and many of the others will be solved in a few years.
They were working on this 20 years ago, so maybe they'll developed it by now.
Obviously, satellite won't have the same resolution through walls as they do with objects out in the open.
By the way, you can see through walls with very high resolution if you are right outside. Three examples:
Incidentally, here's a funny story in which Sony accidentally sends camera made for intelligence purposes which can see through clothes to consumers. Apparently the media actually bought Sony's story that the seeing-through-clothes capability was a total accident.
You can also hear through walls:
"A device known as a laser-guided microphone can be pointed at any pane of glass, allowing the user to eavesdrop upon any conversation emanating from within a windowed structure."Interestingly, the page that you are currently reading was noted as a good place to start for this sort of information by someone who claimed to work on spy satellites as of around 2000: http://answers.yahoo.com/question/index?qid=20091122104929AAmqlS8 . The person also says, "20 years ago we were looking through walls as if they just weren't there, able to take full-color photos of whatever we wanted, limited at the time only by the technology, which is obviously far, far ahead at this point."
btw interesting factoid: while trying to find that article, i found that it's legal for the US govm't to find out what you are watching on TV (if your service provider has that info, i.e. with Tivo, and if they claim it's relevant to a terrorism investigation)
So why does the government want domestic UAVs? I would expect that UAVs would allow even better resolution, cheaper, and more ubiquitously than the spy satellites. Like, maybe the U.S. government would have to concentrate multiple spy satellite resources in order to get a license plate at a particular time, which means that they can only be monitoring so many places at once with the satellites.
Also, UAVs could allow an extremely precise level of resolution, rather than just a decent one, could allow better resolution/more wavelengths when seeing through walls, and could detect audio.
You can see what's on a laptop screen even behind walls:
"I was able to eavesdrop certain laptops through three walls," says Kuhn. "At the CEBIT conference, in 2006, I was able to see the Powerpoint presentation from a stand 25 metres away." Here's the image he managed to get: 
from http://www.newscientist.com/blog/technology/2007/04/seeing-through-walls.html
Also, from Slashdot: Laptops And Flat Panels Now Vulnerable to Van Eck Methods
"Using radio to eavesdrop on CRTs has been around since the 80s, but Cambridge University researchers have now shown that laptops and flat-panel displays are vulnerable too. Using basic radio equipment and an FPGA board totaling less than $2,000 it was possible for researchers to read text from a laptop three offices away."
