[opendtv] Re: More on the 1-pixel camera
- From: Tom Barry <trbarry@xxxxxxxxxxx>
- To: opendtv@xxxxxxxxxxxxx
- Date: Sun, 22 Oct 2006 20:05:20 -0400
The
> camera creates [I would have said 'records' or 'encodes'] an image by
> capturing just one point of light, or pixel, several thousand times ..
I don't at all understand how this could work. It seems there would be
too little information. Whether they capture all those one pixel
samples randomly or in a normal raster scan they are claiming to capture
a megapixel image using only kilopixel samples. Assume for the moment
they did not use their new mathematics to process those samples but
instead stored them to disk. Thus, without any additional compression
(or motion comp) they would already be storing a supposedly good (?)
image with a many hundred to one compression ratio. I'm skeptical.
Am I missing something obvious or are they stretching the "several
thousand times" quantity a bit.
- Tom
Manfredi, Albert E wrote:
This is the news article from Rice University:
http://www.media.rice.edu/media/NewsBot.asp?MODE=VIEW&ID=8912&SnID=60786
67
The concise explanation is stated as:
"In the single-pixel camera, the image from the lens is shined onto a
digital micromirror (DMS) and bounced from there through a second lens
that focuses the reflected by the DMD onto a single photodiode (PD). The
mirrors of the DMS are shuffled at random for each new sample, creating
random patterns of black and white, as depicted in this schematic. Each
time the mirrors shift, a new pixel is recorded by the photodiode. The
camera creates [I would have said 'records' or 'encodes'] an image by
capturing just one point of light, or pixel, several thousand times in
rapid succession. The thousands of single-pixel snapshots are then
assembled by a computer into a high-resolution image."
Okay, so my guess is that the "missing piece" in this explanation is
that somehow that (pseudo-)random sequence used by the DMD in the camera
needs to be duplicated at the decoding end. Then at least it's clear
that information isn't being thrown away.
So my guess is that you need a time sync and knowledge of the
pseudo-random sequence, from camera to display.
Bert
----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:
- Using the UNSUBSCRIBE command in your user configuration settings at FreeLists.org
- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word
unsubscribe in the subject line.
--
Tom Barry trbarry@xxxxxxxxxxx
Find my resume and video filters at www.trbarry.com
----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:
- Using the UNSUBSCRIBE command in your user configuration settings at FreeLists.org
- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word
unsubscribe in the subject line.
Other related posts:
