

		Readers' contributions and Updates
	to the paper "How to find hidden cameras" by Marc Roessler
		     marc@tentacle.franken.de


	The latest version of this file can be found at
	http://www.franken.de/users/tentacle/papers/hiddencam-contrib.txt

	The paper "How to find hidden cameras" can be found at
	http://www.franken.de/users/tentacle/papers/hiddencams.pdf



26 Jul 2002
It seems clear by now that the SpyFinder(tm) works based on the effect
of retro-reflection. Experiments with a red laser pointer showed very visible
reflections from the optics of a standard sized camcorder. Even the lenses
of subminiature pinhole cameras were distinguishable suprisingly well.
In order to achieve acceptable results the sensor (in the most simple case:
your eye) should be as close to the axis of the emitter (e.g. a laser pointer)
as possible. The emitter must be as punctual as possible, a laser
meets this criteria. Obviously this works best at low ambient light levels --
while this all works pretty well at room light level a sunny day will
probably make things more difficult. Because a laser pointer only produces
a small red dot by means of which the whole environment needs to be swept
it is advisable to construct an optical scanner. Mirror motors salvaged
from defective laser printers are a good start. This only projects
a thin red line, so the room still needs to be swept carefully. The scanner
could be made still more comfortable by using a laser module with integrated
optics that project a thin line. Combined with the spinning mirror this will
generate a two dimensional projection by means of which the environment
can be scanned easily.
A drawback of the retro-reflection scan method is that curved objects
made of metal or glass (e.g. bottles) will cause very visible reflections
as well, so extra care needs to be taken to rule out the false positives.


07 May 2002
An anonymous reader not affiliated in any way with SEA Inc. suggested that
the laser based camera detector described in section 4.8 may work based on
retro-reflection. Retro-reflection means the reflection of light right
back to its source, independent of the direction the light comes from [1].
This effect is often noticed on photographs, where the photo flash
can make the subjects' eyes appear to glow red [2,3], caused by the retina
at the back of the eye reflecting the light. Commercial applications
of retro-reflection include traffic signs and reflective clothes commonly
worn by road construction workers. Another application is pupil/eye detection
and tracking [4,5].
The CCD/CMOS chip usually mounted at the focal plane of the video camera's
lens is a good reflector and thus can be expected to cause visible
reflections.
[1] 3M webpage: What is Retroreflectivity
    http://international.3m.com/intl/AU/reflective/retro.html
[2] Cliffshade Digital Photography page:
    Red-eye control
    http://www.cliffshade.com/dpfwiw/red-eye.htm
[3] Photographytips.com: Red eye
    http://www.photographytips.com/page.cfm/434
[4] Antonio Haro, Irfan A. Essa, Myron Flickner:
    Eye Detection and Tracking
    http://www.cc.gatech.edu/cpl/projects/pupil/
[5] Scott Fritzinger, Dr. George Bebis, Dr. Dwight Egbert, Dr. William Wells:
    Infrared Eye Detection
    http://www.cs.unr.edu/CRCD/Summer01/scottf/


25 Mar 2002
Steven Mann pointed me to two of his patents, canadian patent No. 2280420
("Selective vitronic viewing concealed by material such as polymer diffuser")
and canadian patent No. 2354113 ("Intelligent bathroom fixtures and systems").
Both patents are available from the Canadian Patents Database at
http://patents1.ic.gc.ca/. The first patent suggests to use PDLC
(Polymer Dispersed Liquid Crystal) technology for hiding cameras.
PDLC layers are transparent when unpowered, but become intransparent
(frosted look) when AC voltage is applied to them. The patent suggests to
integrate PDLC technology into transparent wall tiles, for example.
Pulsed AC currents are then applied to the assemblys. As long as the tile
is transparent only for very short intervals (this is when the video camera
shutter opens), it seems to be opaque to the human eye and can not be
distinguished from regular wall tiles.
The patent suggests further to integrate cameras concealed by means of this
technology into toilet bowls. This is to detect the presence of humans
sitting on the toilet as well as the presence of waste in the toilet bowl
in order to trigger the flushing mechanism automatically.
(Big Brother in the toilet bowl?)
The second mentioned patent suggests the use of several image sensors
(that is: cameras) for controlling showers, faucets, urinals or water closets
automatically. The cameras are to be integrated into wall tiles, toilet bowls,
the nozzles of showers and similar. In addition the camera images can be
provided to monitoring stations, for security reasons. It is suggested to
"[install this] bathroom control system in a variety of bathrooms in a
bathroom network", providing investigatory access to security officers
or "other officials who may help ensure bathroom security". Transparent bath
tubs are suggested in order to ease monitoring. Mr. Mann considers the
proposed system to "enhance the the privacy of users" by ensuring that
officers and security guards need not enter the showers. In addition the
system provides the benefit of distinguishing "a first class users" from
"a second class of users", providing "a more pleasant shower experience
to a first class of users, and a discomfort mode for providing a less pleasant
shower experience to a second class of users".
See section 3.9, "Taking over pre-installed cameras",
of my paper "How to find hidden cameras".

--- end of document ---
