Vision Flashcards
(227 cards)
What is the sensitivity of the human eye?
It is sensitive to the electromagnetic spectrum between the wavelengths 400nm (violet) and 700 nm (red), with maximal sensitivity at 550 nm (green). The eye can respond to a single photon of light, 5-8 photons arriving within a short time are required to give the experience of a flash of light in the dark-adapted state.
Why is it difficult to distinguish differences in intensity at high light levels ?
Intensity is encoded by the visual system logarithmically
What is the definition of vision?
The process of discovering from images what is present in the world and where it is
What is parallel processing?
The brain uses a 2-D shifting pattern of light intensity values on the two retinas to form a representation of the form of an object, its color, movement and position in a 3-D space. Each of these visual channels are handled simultaneously by distinct but interdependent pathways, to create a unified percept.
What is serial processing
A task is segmented into several subroutines that are executed sequentially.
What does visual processing give higher weight to?
Regions of the world that are changing in time (movement) and space (contrast) than those that are constant
What does visual perception require?
The existence of internal representations of the visual world to allow the brain to make hypotheses about what the retinal image is. Internal representations allow for the fact that vision allows pattern completion and generalisation. Some internal representations are specified during development and are immutable, but most depend on early learning
What is pattern completion?
Generating a complete percept even when the raw sensory data is incomplete or corrupted by noise
What is generalization?
The ability to recognise objects from a wide variety of vantage points and contexts.
What is a visual illusion?
Results from an unresolvable mismatch occurring between the sensory input and the internal representation.
What is perceptual constancy, and why is it important?
Visual perception can be invariant over wide differences in the properties of the retinal image, for eg, color constancy preserves the colors of objects in the face of alterations in the wavelength composition of the light source. Perceptual constancy permits successful object recognition under a wide variety of ambient conditions.
There are both monocular and binocular clues for depth perception, monocular clues are more important for distant objects, where binocular clues cannot be used, what are examples of these?
-Motion parallax- movement of the head causes an apparent movement of near objects with respect to distant ones.
-Geometric perspective- Parallel lines appear to converge with distance
-Relative sizes of objects of known dimensions
-Occultation- Distant objects can be partly hidden by nearer ones
-Extinction- Distant objects are faded and bluer d/t intervening haze
-Ability to resolve fine detail falls off with distance
-Accommodation
What is accommodation?
Neural signals that correspond to how much the visual system has had to change the shape of the lens to keep the object in focus
The binocular clue to depth perception is stereopsis, what is this?
The eyes are 6.3 cm apart, therefore the image of a nearby object falls onto different horizontal planes on the left and right retinas (retinal/binocular disparity). When the eyes converge to fixate on a nearby point the two retinal images of the point are perceived as fused in to a single point. All other points at the same distance are fused. Points in space that lie closer or further away than these, will form images at different corresponding retinal positions and hence different binocular disparities, the closer the object, the bigger the disparity, beyond these, two images are seen (diplopia) or info from one eye is completely rejected by the visual cortex
What is binocular parallex?
Each eye has a slightly different image of the world. When viewing a scene first through one eye and then the other, when nearby objects appear to jump sideways.
With regards to stereopsis, when an object is closer than the close point, images of these points might also fuse, when is the disparity too great for this to happen?
> 0.6 mm or 2 degrees
What is required for stereopsis?
It is only possible for the field of view in which the two monocular visual fields overlap. The brain is able to compute depth from disparity simply by comparing where the same pattern lies on the left and right retinas. It does not require form, movement or colour.
What is the basic definition of colour vision?
It permits boundaries to be seen between regions that have equal brightness, provided the spectrum of wavelengths they reflect is different, but it not just a matter of measuring all the wavelengths in the relected light
What does the spectrum of light reflected from an object depend on?
On the wavelength composition of the illuminating light and the reflectance of the surface
What is dichromatic color Vision?
It requires a minimum of 2 types of receptor, that respond over different wavelength ranges. With these receptors, the visual system can assign 2 brightness values for each pixel of the visual field. By comparing these values, colors may be perceived. If a pixel reflects more short- wavelength light it will appear brighter to a short- wavelength receptor than a long wavelength receptor and will be seen as blue. If a pixel reflects reflects more long- wavelength light it will be seen as red. If a pixel reflects equal amounts of short and long wavelength light - it will appear monochrome, either white or shades of grey depending on intensity of light
Human color division is trichromatic, why is called this?
The human eye has 3 populations of receptors (cones) that function is daylight, each sensitive to a different range of wavelengths. The system abstracts 3 brightness values for an object and comparisons of these determine color
How do cones transmit the color of objects?
The 3 types of cones have maximum absorptions corresponding approximately to violet, green and yellow light. The wavelength of the light does not affect the character response of the cone. A given cone simply has a higher probability of absorbing a photon which is closer to its peak wavelength. The visual system has no way of detecting the absolute wavelength composition of any light.
What is perceptual cancellation?
Some colors in the same pixel of visual space perceptually mix to produce other colour categories, but complementary colors do not perceptually mix
what is simultaneous color contrast?
Is the perceptual facilitation of complementary colors that occurs across boundaries eg a gray disk within a red background looks slightly green