A little more about the octagonal CCD pixel - which gives an apparent
result of double the actual pixels. Some people claim it to be an
alternative form of interpolation some don't.
In general, you cannot sample any finer than your sampling aperture,
which is the individual cell size. (I think this was figured out by
Shannon.)
By moving cells (or moving your source material) an increment of the
cell size, you in effect average half of two adjacent cells, but the
result is no different than if you did it in software, except slower.
Manufacturers should (IMHO) call this "mechanical interpolation,"
rather than mis-labeling it as optical resolution.
I don't understand how a hexagonal array would provide higher
resolution, unless they are mechanically shifting the array somehow,
which would then simply yield mechanical interpolation.
I suspect the hexagonal array allows them to better use a given area
of silicon, thus improving yield and lowering cost,. Then some bright
marketing guy and some engineer got together and figured out a
sophisticated way to exploit the hex array so they could lie about
resolution.
But I could be wrong. Maybe someone who has studied signal processing
and information theory closer than 25 years ago has better insight.
Tim?
: Jan Steinman <mailto:Jan@xxxxxxxxxxxxxx>
: Bytesmiths <http://www.bytesmiths.com>
< This message was delivered via the Olympus Mailing List >
< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >
|