Biomedical Engineering Reference
In-Depth Information
There are at least 10 known variants of animal
eyes
[1, 28]
. These 10 types can be grouped as
either noncompound eyes or compound eyes. Of
the noncompound eyes, the most frequently
mimicked type for vision sensors is the
refractive
cornea eye,
which is often called a
camera eye,
since
it is very similar to how most cameras operate.
This is the type of eye found in almost all mam-
mals, birds, reptiles, and most other terrestrial
vertebrates. For completeness, we discuss this
type of eye briefly in the next section. Of the
compound eyes, the three most commonly
mimicked types for vision sensors are the
apposi-
tion eye,
the
optical superposition eye,
and the
neural
superposition eye.
We cover these types of eyes in
more detail, since some of the most recent work
in biomimetic vision sensors is based on them.
L
R
Iris
F
V
O
C
O
O
FIGURE 1.12
Simplified cross-sectional diagram of the
human eye.
(necessary for good static acuity) in the short
focal distance required of a compact vision
organ.
15
An artificial vision sensor based on a
camera eye typically uses a single large-aperture
lens or lens system (mimicking the cornea and
lens) combined with a relatively large, high-
resolution focal plane array of photodetectors
(mimicking the photoreceptors in the retina).
The human eye is a highly complex sensor
that responds to electromagnetic stimuli at
wavelengths of approximately 400-700 nm
[29,
30]
; for obvious reasons this band is called the
visible wavelength
. Ambient light enters the cor-
nea and through the anterior chamber (contain-
ing the aqueous humor), through the pupil
opening of the iris (which determines the effec-
tive aperture size), and through the crystalline
lens and then passes through the vitreous humor
before striking the retina. The retina consists of
many layers of neural tissue that contains,
among other things, the photoreceptors (rods,
cones, and nonimaging photosensitive ganglion
1.3.1 Camera Eye
Nearly all mammals, including humans, have
camera eyes. As an example,
Figure 1.12
is a
highly simplified diagram of the human eye.
The region between the cornea and the lens
(the anterior chamber) is filled with a waterlike
substance called the
aqueous humor
; the region
behind the lens (the posterior chamber) is filled
with an optically clear but somewhat gelatinous
substance called the
vitreous humor
(or
vitreous
body
). In this type of eye, the primary refractive
power is due to the air/cornea optical inter-
face.
14
An additional refractive effect is some-
times provided by an internal lens, such as the
variable-shape crystalline lens that humans use
to accommodate focus for close objects.
The use of significant refractive power allows
the use of a relatively large aperture in the
camera eye, permitting good light gathering and
keeping the blur spot acceptably small
14
Aquatic mammals must use a lens as the primary refractive element, since the index of refraction of water is very
close to the indices of refraction of the cornea and the aqueous humor.
15
Due to aberrations in the human eye, the actual PSF is somewhat larger than the diffraction-limited Airy disk and has an
approximately Gaussian shape. Aberrations are greater, and therefore the PSF is wider, for larger pupil diameters.
