The currently used image stabilization methods use either electronic
sensing or mechanical (gyro) sensing. They use either electronic or
electro-optical image correction.

Electronic sensing involves a frame store buffer and a microcomputer or
DSP that 'locks' onto details in the image. When they shift it senses the
direction and magnitude of the shift. This is the earliest system that was
used, by Panasonic I think, more than 6 years ago afaik. It is confused by
rotation combined with shifting.

Mechanical sensing involves rate gyros with no moving parts made by Murata
or Tokin. They use the Coriolis effect as someone else has pointed out.

Electronic correction involves a frame buffer that has a readou window
that is movable in it. In other words, only about 80% of the real image is
output, and the controller shifts this smaller rectangle in the larger
image to compensate (apparently) for shake and motion.

Optical correction involves an electro-optical system that 'bends' the
lightpath in the camera under control from the computer. This is the best
system on sale today afaik, and this is what is called 'Optical image
stabilization'. It usually uses ceramic gyros and is independent of a
digital image of any kind. So it works in binoculars and still lenses. The
major deployers are Sony and Canon.

The best systems on sale today combine both feature sets for outstanding
results. You can shoot a picture of a 3 ft object at over 100 meters using
image-filling zoom freehand (no tripod, nothing). Of course I suspect that
military systems are FAR batter than this. Especially them things the
Comanche helicopters carry under their noses (and which things seem to
have been adapted for Olympics use in Sydney - or were those their poor
relatives ;-).

Peter

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu