| |
|
|
|||||
| |
![]() |
![]() |
|
||||
![]() |
![]() |
![]() |
|
||||
![]() |
|
||||||
![]() |
|
||||||
![]() |
|
||||||
![]() |
|
||||||
![]() |
|
||||||
| |
![]() |
|
|||||
![]() |
|
||||||
| |
|
||||||
| |
|
||||||
| LIGHT
AND CAMERA INTERACTION CONCEPTS BASIC CONCEPT: WITHOUT LIGHT,
THERE IS NO IMAGE. The proper
choice of video cameras
and lenses can extend that range down into the ultraviolet (below 400
nm) and
up into the near infrared (above 760 nm). For special applications,
special
imagers with sensitivity in the range of 7 to 15 microns(7000 to 15000
nm) are
available to image emitted (sentient) body heat of warm blooded animals
when
there is sufficient thermal contrast with the background. CCD (Charge
Coupled Device) video
cameras are inherently more sensitive to infrared light than visible
light. Therefore
color cameras must
filter out (or cut) infrared light to properly balance the visible part
of the
spectrum to create a color image.
Color
cameras are available that can image the full visible color spectrum
under
white light conditions and when required, physically remove the
infrared cut
filter from the light path allowing sensitivity into the infrared as
well as
any remaining visible light. This
function will usually increase the camera sensitivity by a factor of
10. Monochrome
cameras are inherently
more sensitive and have higher image resolution than color cameras
because each
pixel is sensitive to the full spectrum of light. Color cameras require
selective filtration (red, green, blue) over each pixel to proper
render a
color image. Cameras are basically pretty dumb. Simple cameras interpret the world as an all gray background with 18% reflectivity (average reflectance of a Caucasian face). Therefore if you point it at a white surface, the camera reduces the amount of light falling on the imaging chip by increasing the electronic shutter speed (or reducing iris size) and outputs the image as 18% gray. If you point it at a black surface, the camera increases the amount of light falling on the imaging chip and outputs the image as 18% gray. Now you know why simple cameras cannot properly image small areas of interest when 80 to 90% of the imaged background is either very bright or dark. You have to move in close. A simple camera compensates over a light to dark image area ratio of 8 to 1. Above 8, everything is white, below 1, everything is black, 7 to 2 are shades of gray. Technology to
the rescue
(partially). A new
generation of
cameras with Digital Signal Processing (DSP) can be manually programmed
to
interpret the light being reflected off of only a portion of the image
area
while ignoring the glare or shadows in other area of the image. It is
generally
referred to as Back Light Compensation (BLC) from it origins in the
security
industry. The very
latest generation of
cameras can automatically perform DSP on every pixel in the image area
thus
increasing the dynamic light compensating ratio to 128 to 1. However,
it still
cannot compensate for fast moving reflections like ripples on a water
surface
because of damping factors built into the DSP system. LIGHT "WHITE"
LIGHT All "white
light" is not
the same especially when viewed by a color camera.
Portions of the spectrum may be missing. For example
White Light
Emitting Diodes (LEDs) are actually a combination of a blue LED emitter
under a
phosphor coating that when "excited" by the blue light and filtered
by the coating create a "white light" as interpreted by the human
eye. Likewise
fluorescent lights appear
to be white but the phosphors can be biased to provide warm light, cool
light,
"natural" light or daylight spectrums but they are not the continuous
spectrum which incandescent lights provide. An additional problem with
line
powered (60 HZ) fluorescent lighting is that it is slightly out of
synchronization with a camera's base frequency.
This causes a color
image
to slowly alternate between a cool image (bluish) and a warm image
(reddish). Certain
cameras can cancel
out this fluctuation while other cameras require special high frequency
(20
KHZ) fluorescent lights. Outdoor "white"
daylight
is also not constant. Warm early morning light may have a spectrum
centered
around 2400º Kelvin while noon daylight with a blue sky may be
12000+º
Kelvin. Your eyes
compensate automatically
but a camera requires "white balance compensation" if colors are to
be recorded accurately. Infrared
illuminators are, in
general, the spectrum of choice when humans and other vertebrates are
to be
observed covertly and not bias the viewing of those subjects. In
addition,
infrared illumination will not attract unwanted attention by
non-targeted
subjects and predators. However, IR illumination can easily be observed
with
various "night vision" devices. For most
applications, infrared
Light Emitting Diodes (LEDs) provide sufficient illumination to
distances of
150 ft. (50 m). For extreme distances, long pass filtered (far red and
infrared
pass through, visible is absorbed) incandescent line powered (120 VAC) illuminators are available. If these tiny
"eyes" are a
potential problem, the solution is use LEDs with peak emission in the
940 to
950 nm range. With
these emitters there
is no "spillover" into the visible spectrum and they are totally
covert for all know vertebrates. If
the
emitter is very close to the subject, say 1 ft. (30 cm) or less
depending on
the number of emitters, some low level heat (far infrared being
radiated by the
LED housing) can be felt but no visible light can be seen. For underwater
observations in the
infrared, the 840 to 880 nm are the emitters of choice because their
spectrum
is not as rapidly absorbed by the water and they are generally usable
to 3.3
ft. (1 m). The 940
to 950 nm LEDs can
also be used, if required, but the working distance is even shorter. TECHNOLOGY AND TOOLS FOR THE 21ST CENTURY™ |
|
©Copyright
2007, Fuhrman Diversified, Inc.
|