Loading ...
Sorry, an error occurred while loading the content.

Re: Digital camera "eye" image?

Expand Messages
  • feedbackdroids
    ... not], ... layer. ... rolled up ... faced the ... surface, ... vertebrate, it ... migrate ... first ... This layer ... sensed. When ... inside of the ...
    Message 1 of 14 , Apr 1, 2006
      --- In ai-philosophy@yahoogroups.com, "ray scanlon" <rscan@...> wrote:
      > feedbackdroids wrote:

      > > Actually [and putting aside the issue of intelligent design, or
      > > the retina is built inside out because the photoreceptors have an
      > > enormous energy requirement, which is supplied by the choroid
      > I would put it another way. When the sensory net of a (flatworm?)
      rolled up
      > and went inside (making the first vertebrate), the eyespots then
      faced the
      > interior of the neural tube. When the eye cup migrated toward the
      > the retina remained facing the interior. If you wish to be a
      vertebrate, it
      > is necessary to put up with things like that
      > The cephalopod eye forms on the surface and the optical axons
      > inside. The eye is right side to. How nice!
      > > Also, in nocturnal animals, photons which miss the rods on the
      > > pass are reflected from the closely-positioned pigment epithelium
      > > back to the rods, which improves vision at low light levels.
      > Don't think so. The flatworm eye is covered by a pigmented layer.
      This layer
      > transduces light energy to heat energy. Heat energy is easily
      sensed. When
      > this eye went inside, the pigmented layer ended up facing the
      inside of the
      > neural tube. It is not there to improve low-light vision. It is
      just there,
      > like the appendix is just there. Evolution can be messy..
      > Ray

      I don't know much about the "insies and outsies" of flatworm vs
      vertebrate evolution, and I was mainly addressing whether the
      inverted design is really a blunder, or rather a good solution from
      an engineering perspective.

      It seems most of the accessible info on vertebrate retinal design is
      on ID sites [maybe I am reading too many of those], but here is what
      one article says .... I'm afraid they hooked me when they mentioned
      Helga Kolb's name ...

      The photoreceptors (rods and cones) must also face away from the
      front of the eye in order to be in close contact with the pigment
      epithelium on the choroid, which supplies the photoreceptors with
      blood. This arrangement allows a "steady stream of the vital molecule
      retinal" to flow to the rods and cones without which vision would be
      impossible (Kolb 2003, p. 28). The verted design, claimed by Miller
      to be superior, would place the photoreceptors away from their source
      of nutrition, oxygen, and retinal (the choroid). This design would
      cause major problems because rods and cones require an enormous
      amount of energy for their very high metabolism required in
      functioning, maintenance, and repair. In addition, because of
      phototoxicity damage, the rods and cones must completely replace
      themselves approximately every seven days or so.

      They do mention heat dissipation too ...
      The photoreceptors and retinal epithelium absorb an enormous amount
      of light on a continuous basis when the eyes are open. Because the
      light is converted largely into heat, the retina must have a very
      effective cooling system, again provided by the choroidal blood
      supply directly behind the pigment epithelium. If the pigment
      epithelium tissue were placed in front of the retina, sight would be
      seriously compromised. Reversing the retina so that it faces away
      from the pigment epithelium would also compromise sight to the degree
      that sight would be impossible because the photoreceptors must be
      embedded in the retinal pigment epithelium to obtain the nutrients
      required to function.

      Regards scattering off neural elements, they mention ...
      In the retina region which has the highest resolution, the central
      retina (the fovea and, in particular, the foveola), the neurons in
      front of the photoreceptors are shifted to the side so that light has
      a direct pathway to them, resulting in the least distortion where it
      matters most.

      Regards reflected light off the pigmented layer, they mention ...
      .... the retinal pigment epithelium (Martínez-Morales 2004, p. 766).
      This monolayered tissue contains the black pigment melanin that
      absorbs most of the light not captured by the retina. This design has
      the very beneficial effect of preventing light from being reflected
      off the back of the eye onto the retina, which would degrade the
      visual image.

      However, certain noctural animals do have an additional layer of
      tissue covering the pigment epithelium called the "tapetum", and
      which does reflect light back to the photoreceptors. Witness how a
      cat's eyes "glow" in the dark when a light beam is shined on them,
      while humans do not glow. Similarly, the eyes of pesky racoons glow,
      such as were living under the shed this winter, and won't go away.

      In many species for whom vision in very low levels of illumination is
      important, a layer of reflective crystalline material, the tapetum
      (Latin: carpet) is incorporated in the RPE or choroid.[3] Acting as
      a mirror, the tapetum reflects light which has passed between the
      photoreceptors, so augmenting the light bombarding the
      photoreceptors. Hence the proverbial `cat's eyes' when caught by a
      beam of light in the dark.
    • feedbackdroids
      ... a ... Given my proclivities, I very much like the behaviorist tinge you put on this post ;-). ... differences. ... Whatever this means. ... As indicated,
      Message 2 of 14 , Apr 1, 2006
        --- In ai-philosophy@yahoogroups.com, "John J. Gagne"
        <fitness4eb@...> wrote:
        > --- In ai-philosophy@yahoogroups.com, "John J. Gagne"
        > <fitness4eb@> wrote:
        > >
        > > Certainly, there's a significant difference between the design of
        > > digital camera's CCD array (or CMOS if you prefer) and the human
        > > eye's retina array. While I'm sure there are many differences we
        > > could discuss, the one I'm most interested in can basically and
        > > simply be stated as, "The retina does not employ the RGB-
        > > triad/pixel method of sampling light".
        > >
        > > Question #1 Why not?
        > Because "seeing" doesn't happen in snapshot frames. Seeing is not
        > about capturing high resolution snapshots. It's about measuring how
        > and when change happens over time.

        Given my proclivities, I very much like the behaviorist tinge you put
        on this post ;-).

        > >
        > > Question #2 Why this crazy layout?
        > This crazy layout of the retina is optimized for finding

        Whatever this means.

        > > Q3 Why not triads/pixels all over the retina?
        > Because it's simply not necessary for the purposes of "seeing". I
        > think Chris hit the nail on the head when he said something like
        > seeing can't be separated from processing.

        As indicated, the brain moves the eye such that the fovea is centered
        upon the region of interest in the visual field. And the tiny part of
        the visual field that corresponds to the fovea is greatly magnified
        in its mapping/representation in the visual areas of cortex.

        This clearly turns out to be an engineering compromise from an
        evolutionary perspective. The brain just doesn't have enough
        processing power to perform the same degree of processing on each and
        every pixel in the visual field, so evolution has traded off temporal
        for spatial. IE, high-res processing on a small area of visual space,
        with temporal scanning to build up an internal representation, which
        no doubt requires a certain amount of short-term memory buffering in
        cortical areas in order to create a coherent internal image.

        > > Q4 If we built a digital cameras CCD array with this type of
        > > arrangement, what would a snapshot image look like?
        > I think the gray scale image would be really good... But not enough
        > information to render any color! But, if we cheated and manually
        > colored in the dots at the correct intensity I think we could get a
        > fairly good idea from such a CCD.
        > > Q5 What type of processing algorithm could transform the image
        > > from our modified CCD in Q4 into a photo-realistic triad/pixel
        > > image
        > None! None! None! The necessary information to render color is just
        > not there! Difference is something different though... ;o)

        Whatever this means.

        > > and how fast would the computer have to be to transform, 24
        > > per second (or whatever the rate of persistence is) of such
        > As to this second part of Q5, I don't really know but I think this
        > is an important question because it involves/is intimately related
        > to the processing part.
        > I think the term "resolution" has a very different meaning with
        > respect to seeing then it does for photography. For a digital
        > camera, resolution is all about number pixels. For a "visual array"
        > it's about number of samples over time.
        > Chris also mentioned how the eye is in continual motion when
        > examining the field of interest. This makes perfect sense
        > if "difference" defines "interest".

        Yes. Due to the spatial vs temporal tradeoff mentioned above, which
        is no doubt related to the brain's having only a limited amount of
        internal processing power, it must temporally scan the hi-res fovea
        to bear upon areas of interest in the visual field. No doubt the only
        reason the brain can make any sense of all this is due to short-term
        memory buffering in the cortical regions.
      Your message has been successfully submitted and would be delivered to recipients shortly.