Cameras with more and more pixels are common. But the latest research
is revolutionizing the technology by using cameras with just a single
pixel. See the URL and excerpts below. (01)
This R & D is important for practical applications. But it also helps
us understand how the brain can create a unified, stable 3D image of
a scene from multiple pairs of 2D images from eyes that are constantly
flitting from one part of the scene to another. (02)
Furthermore, each eye only has high resolution for a small spot in
the center. Yet we can "look at" a fixed spot and get the impression
a stable image, even though the eyes are in motion. (03)
There is a huge amount of computation going on in the brain. Research
on biological systems and on engineering methods is complementary.
Insights from either one suggest new directions for research in the
other. (04)
John
________________________________________________________________________ (05)
http://www.technologyreview.com/view/520006/first-3d-movies-from-a-single-pixel-camera/ (06)
First 3D Movies From a Single Pixel Camera (07)
Single pixel cameras are taking the world of imaging by storm. These
counterintuitive devices have the ability to photograph an entire scene
in 3D and at a resolution of choice using a single pixel. Some versions
do not even need a lens. These resultant images are entirely free of the
optical aberrations that lenses can introduce; indeed the entire scene
is always in focus... (08)
Single pixel cameras rely on a technique known as compressed sensing.
The idea is to pass the light from a scene through a medium that
randomises it and then focuses it on to a single pixel. This randomising
medium could be a piece of frosted glass, a spatial light modulator or,
as in this case, a digital micro-mirror device in which the mirrors are
arranged at random... (09)
The signals picked up by the single pixel may seem random but in fact
they are correlated because they all come from the same source–the
original scene. (010)
So the trick behind compressed sensing is to analyse the data in a way
that finds this correlation. Once the correlation is known, it is
relatively straightforward to reconstruct the original scene. (011)
That produces a simple 2D image with a resolution that depends on the
number of single pixel samples that have been taken. (012)
Making 3D image is straightforward too. For this, the scene must be
illuminated by a laser light and the single-pixel circuitry gated so
that it is sensitive for a short instant after the illumination, when
the light bounces back. (013)
In this way, it is possible to sample light that returns from a specific
distance from the camera. And by changing the timing of the gating, it
is possible to build up an entire 3D image. (014)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (015)
|