AirStrike is a gesture control system designed to enable control of any pc
or display interface via waving your fingers in the air, and smart sensing technology
to detect those gestures accurately enough to completely replace a computer
mouse or 3D pointer. It is not a display technology itself, simply the ability
to control displays.
It was born in part, by a desire to prove that touchscreen was not the only
way to go. A system that worked like touchscreen but without actually touching
the display, would work even where there was no screen to touch; head mounted
displays and holograms.
Because it works off of both visible light and the infrared range, it is capable
of controlling a liquid droplet hologram system such as Fogscreen or Heliodisplay,
where other technologies have failed. This is because water droplets are invisible
to infrared light, and do not detract from the detectable finger movements.
Airstrike detects finger movements by way of one or more input strips embedded
with two CCDs each. For a PC, that input strip is a normal keyboard with a CCD
at either end, for triangulation purposes. However, in larger displays, for
general public use, the keys can be foregone in favour of a featureless strip
bar sensor.
CCD stands for Charge-coupled device. The type of CCD used by Airstrike is designed
to capture and process images, and is functionally identical to the sensor type
found in digital cameras. It works by detecting light intensities in its determined
optical range. On the top of the CCD there is a photoactive region defined by
an array of light sensitive capacitors and a transmission region directly under
it.
An image is projected by a lens on the photoactive region, at a rate of several
dozen times per second. Each time this occurs, the capacitors electrical charge
alters proportional to the light intensity at that location in the snapshot
of the world it views. This is essentially a light intensity greyscale digital
photograph which the transmission region converts into a pulse stream that can
be understood readily by machine vision systems. Two of these CCDs on a bar
about 30 cm apart, create two different images of the same area in near-space,
which can readily be combined into a stereoscopic image used for exact spatial
placements of a user's fingers in 3D space. From this, and from changes in finger
position over very short spaces of time, the desired command is inferred and
acted upon.
Commands are simple and intuitive. In general the user's finger takes the place
of a mouse cursor, and pressing that finger towards the screen signifies a click.
However, most interestingly the system is fully capable of using all the fingers
on both hands at the same time, to interpret an instruction or set of instructions.
That is, as long of course, as the mass of one hand is not obstructing the other
from the two CCDs. More units are necessary for a true 3D picture to be constructed,
but for most uses, two suffice.
Once an item is selected on the screen, the gesture of grabbing at something
in space and pulling it apart or unfolding it, does just that - it zooms an
image in, filling up the screen. Virtual environments, paint programs, even
complex menu systems can be accessed in this way, without the necessity for
a computer mouse at all.
Hopefully we will see more such gesture control systems in the near future,
competing with touchscreens.
Bookmark
