h a l f b a k e r y
carpe demi
idea:
add, search, annotate, link, view, overview, recent, by name, best, random
meta:
news, help, about, links, report a problem
account:
Browse anonymously,
or get an account
and write.
or Create a new account.
|
|
|
Given that physical surfaces reflect light, such surfaces can be thought of as mirrors of variant levels of perfection. On a reflectiveness scale, a true mirror would be near the top, a glass wall would be pretty reflective and a brick wall would appear to be non-reflective at all.
However, though
it appears to be non-reflective, clearly some indirect light is reflecting off it, just like it would be of a mirror wall. It is just too scatered by the surface.
If the wall was a mirror, our eyes would be able to see both the mirror, and the objects around the corner, reflected in the mirror.
The proposed device would be able to analyze the light coming off an arbitrary surface in much the same way.
When pointed at the the brick building at the corner, it would filter out the direct reflection photons, which are describing the wall itself, but keep and amplify the scaterred photons that are desribing the objects around the corner, just as if the brick wall was in fact made of a highly reflective surface.
[link]
|
| |
So is this a goal or a principle? Need an idea. How does this differ from a mirror? What visual light are you trying to filter out? What sort of analysis would be preformed? |
|
| |
So many questions, so sparse an idea... |
|
| |
Let me attempt to clarify... |
|
| |
Interesting, thanks for the clarification, but sounds like magic unless you can describe how to determine the ordering of randomly scattered photons. |
|
| |
I can see this working if the wall were covered with an array of small reflectors, pointed in random (but deterministic and known) orientation. A person looking at the wall would just see a mishmosh (that's a technical term,) but the r-oscope could be made aware of the pattern and rearrange to create a sensible image. |
|
| |
csea, I was thinking of a dark camera chamber with a filter that's taking out direct light but registering side intensity side light. |
|
| |
Information and energy are easily separable. This information can not be retrieved in the way you describe. no. |
|
| |
//This information can not be
retrieved// |
|
| |
I dunno. Here are two arguments in
opposite directions. |
|
| |
a) For. Consider a surface made of two
mirrors which are not co-planar (a
'broken mirror'). You could easily
"decode" the image reflected from such
a surface, based on (i) knowledge of
what an image should look like or (ii)
the use of a reference image reflected
by the surface or (iii) detailed physical
measurements of the surface. Now
consider a surface made of 100 non-
co-planar reflectors. Same thing
applies, but it's more difficult. Now
consider a surface made of 1 million
non-co-planar reflectors, etc etc. The
problem becomes more difficult, but at
no point completely intractable. |
|
| |
b) Against. Light only humours you by
appearing to travel in straight lines and
reflecting at an angle equal to the
incidence. In fact, it does all sorts of
weird stuff at the quantum level. The
overall effect of light travelling in
straight lines etc only works for
relatively broad light paths, where all
the possible paths can interfere such
that the "classical" straight path
dominates. In the case of a sufficiently
rough surface, this ceases to be the
case, and hence light cannot be treated
as if it travels in straight lines, reflects
at an equal angle etc. Yet, without
these assumptions, you can no longer
reconstruct the image from the
reflection. So, you're screwed at the
quantum level. |
|
| |
I suspect (b) is right. However, if it's
any consolation, I think it ought to be
possible to develop a camera that can
deconvolute the distorted images seen
through patterned (though not frosted)
glass. |
|
| |
Yeah, this is an interesting idea, but impossible on any but the simplest surfaces. You'd need to do some kind of X-ray mapping radar on the surface and then then a receiver large enough to collect light from all the reflected angles. Then a computer large enough to reassemble the data into an image. |
|
| |
But the piece you are missing is the unknown absorbsion of the surface. If the surface is orange, then it will only reflect light in the orange bands and absorb all others. If you are really unlucky then the surface is black and will radiate all the light in the infrared. |
|
| |
I think the light source must be stable and well characterized, as well. |
|
| |
Consider the scenario consisting of a figure to be imaged (such as a standing human) who is back-lit with our favorite outdoor light source, the Sun. |
|
| |
Fortuitous timing, weather, season, etc., could cause a shadow to be projected on the wall so that it would be visible from our observer's location around the corner. On a cloudy day, nighttime, etc., all bets are off. |
|
| |
Given the supposed symmetry of the situation, on a different day and sun location, subject and observer might well be transposed! |
|
| |