DIGITAL IMAGE PROCESSING (IN THE AI ERA)
Xiangyu Yu School of Electronic and Information Engineering, South China University of Technology, P. R. China [email protected] LECTURE 2 THE HUMAN VISUAL SYSTEM “Seeing is believing”
3/10/2020 DIGITAL IMAGE PROCESSING 3 CONTENTS
The human visual system Image Formation in the Eye Illusions
3/10/2020 DIGITAL IMAGE PROCESSING 4 INTRODUCTION
In many image processing applications, the objective is to help a human observer perceive the visual information in an image. Therefore, it is important to understand the human visual system. The percentage of information that flows through visual pathways has been estimated at 90–95% for a typical human without any sensory impairment. The human visual system consists mainly of the eye (image sensor or camera), optic nerve (transmission path), and brain (image information processing unit or computer). It is one of the most sophisticated image processing and analysis systems.
3/10/2020 DIGITAL IMAGE PROCESSING 5
Vision
2017)
( (3e) (3e)
CVIPtools THE HUMAN VISUAL SYSTEM
: Human and Computer Computer and Human :
Analysis
Applications with with Applications
Digital Image Processing and and Processing Image Digital
, ,
Umbaugh from Scott E E Scott from
Figure The Human Visual System Images taken taken Images 3/10/2020 DIGITAL IMAGE PROCESSINGUmbaugh3E6 -P.360 HUMAN VISUAL SYSTEM
The best vision model we have! Knowledge of how images form in the eye can help us with processing digital images We will take just a whirlwind tour of the human visual system
3/10/2020 DIGITAL IMAGE PROCESSING 7
STRUCTURE OF THE HUMAN EYE 2008) Cornea and sclera keep everything in! The choroid contains all of the blood vessels that serve as nutrition to the eye. The iris contains the pigment that gives our eyes their colour The iris diaphragm controls amount of light that enters the eye.
Figure Simplified diagram of a cross section of the human eye.
3/10/2020 DIGITAL IMAGE PROCESSING G3C-P.20-21 Images taken from Gonzalez & Woods, Digital Image Processing ( Processing Image & Digital Woods, fromGonzalez taken Images G3E-P.58-59 STRUCTURE OF THE HUMAN EYE(2)
The lens and the ciliary muscle focus the reflected lights from objects into the retina to form an image of the objects. The lens contains 60-70% water, 6% of fat. The lens focuses light from objects onto the retina
3/10/2020 DIGITAL IMAGE PROCESSING 9 G3C-P.21 G3E-P.58-59 RETINA
Retina: consist of receptors Pattern vision is afforded by the distribution of discrete light receptors over the surface of the retina. There are two classes of receptors: cones (light sensors) and rods. Rods for general vision, cones for details.
3/10/2020 DIGITAL IMAGE PROCESSING 10 G3C-P.21 G3E-P.59 LIGHT RECEPTORS IN THE RETINA
About 6-7 millions cones for bright light vision called photopic - Density of cones is about 150,000 elements/mm2. - Cones involve in color vision. - Cones are concentrated in fovea about 1.5x1.5 mm2. About 75-150 millions rods for dim light vision called scotopic - Rods are sensitive to low level of light and are not involved color vision.
3/10/2020 DIGITAL IMAGE PROCESSING 11 CONES
There are 6 to 7 million cones in each eye. Concentrated in the central portion of the retina called the fovea. Highly sensitive to color. Women have extra cones which means they can see more colors Each cone is connected to its own nerve end, so human can resolve fine details. Cone vision is called photopic or BRIGHT-LIGHT VISION
Photopic (bright-light) vision: vision with cones color receptors, high resolution in the fovea, less sensitive to light
3/10/2020 DIGITAL IMAGE PROCESSING 12 G3C-P.21 G3E-P.59 RODS
There are 70 to 150 million cones in each eye. Rods are more spread out and are sensitive to low levels of illumination Distributed over the retina surface. Several rods are connected to a single nerve end. Rods don’t discern fine details. Rods give a general picture of the field of view. Give overall picture with reduced detail. Not involve in color vision and sensitive to low levels of illumination. Rod vision is called scotopic or DIM-LIGHT VISION. Scotopic (dim-light) vision: vision with rods color blind, much more sensitive to light (night vision), lower resolution
3/10/2020 DIGITAL IMAGE PROCESSING 13 G3C-P.21 G3E-P.59
2002)
( Inside(3e) Inside(3e)
Figure Sensitivity of rods and cones based on measurements by Wald. 3/10/2020 DIGITAL IMAGE PROCESSING 14
Images taken from William K. Pratt , Digital Image Processing: PIKS PIKS Processing: Image K. Pratt, from William taken Digital Images Pratt3E-P.27
Vision
2017)
(
(3e) (3e)
CVIPtools
: Human and Computer Computer and Human :
Analysis
Applications with with Applications
Digital Image Processing and and Processing Image Digital
, , Umbaugh
Figure Relative Response of Rods and Cones. (a) Rods react in low light levels, scotopic from Scott E E Scott from vision, but respond to only a single spectral band, so cannot distinguish colors, (b) Cones react only to high light intensities, photopic vision, and since there are three different types
that respond to different spectral bands, they enable us to see color. The response Images taken taken Images functions3/10/2020 of the cones are the called the tristimulus curves, for “three stimuliDIGITAL IMAGE.” PROCESSINGUmbaugh3E15 -P.362 HEXAGONAL PIXEL
Cone distribution on the fovea (200,000 cones/mm2)
•Models human visual system more precisely •The distance between a given pixel and its immediate neighbors is the same •Hexagonal sampling requires 13% fewer samples than rectangular sampling •ANN can be trained with less errors
3/10/2020 DIGITAL IMAGE PROCESSING 16 MORE ON THE CONE MOSAIC
The cone mosaic of fish retina http://www.nibb.ac.jp/annual_report/2003/03ann502.html
Lythgoe, Ecology of Vision (1979)
Human retina mosaic The mosaic array of -Irregularity reduces visual acuity for high-frequency signals most vertebrates is -Introduce random noise regular 3/10/2020 DIGITAL IMAGE PROCESSING A MOSAICKED MULTISPECTRAL CAMERA
3/10/2020 DIGITAL IMAGE PROCESSING 18 RETINAL PHOTORECEPTORS
•Fovea : Circular indentation in center of retina, about 1.5mm diameter, dense with cones. •Photoreceptors around fovea responsible for spatial vision (still images). •Photoreceptors around the periphery responsible for detecting motion. •Blind spot: Point on retina where optic nerve emerges, devoid of photoreceptors. •Blind spot is the region of emergence of the optic nerve from the eye. •In the blind spot there are no receptors. •1.5 mm 1.5 mm square contain 337000 cones 5mm 5mm CCD imaging chip G3C-P.20-21 3/10/2020 DIGITAL IMAGE PROCESSING G3E19 -P.59-60 CONES VS. RODS
3/10/2020 Figure Distribution of rods and cones in the retina. DIGITAL IMAGE PROCESSING 20 DISTRIBUTION OF RODS AND CONES IN THE RETINA
Degrees from visual axis (center of fovea) Except for the blind spot, the distribution of receptors is radially symmetric about the fovea. G3C-P.21 3/10/2020 DIGITAL IMAGE PROCESSING G3E-P.59-60 BLIND-SPOT EXPERIMENT
Draw an image similar to that below on a piece of paper (the dot and cross are about 15 cm apart)
Close your right eye and focus on the cross with your left eye Hold the image about 50 cm away from your face and move it slowly towards you The dot should disappear!
3/10/2020 DIGITAL IMAGE PROCESSING 22
EYE OF OTHER ANIMALS 2011) ( Spider eyes consist of multiple sets with many individual lenses and sensors, producing comparatively low resolution but broad coverage. Bird eyes have high acuity and resolution. The chameleon can swivel its eyes independently to track different objects in left and right visual
Processing Handbook(6e) Handbook(6e) Processing fields. The adult flounder has both eyes on one side of its head. The eyes of the
Image Image octopus have very good color sensitivity but evolved with the neural circuitry on the opposite side of the retina from mammals. Primates are well adapted for Russ, The The Russ, stereo vision and have greater sensitivity to red colors than most other mammals.
R6C-P.56 Images taken from taken Images 3/10/2020 DIGITAL IMAGE PROCESSING 23 R6E-P.87
WHAT THE EYE TELLS THE BRAIN
2011)
(
Processing Handbook(6e) Handbook(6e) Processing
Image Image Russ, The The Russ,
Figure The principal layers in the retina. Light passes through several layers of processing neurons to reach the light-sensitive rod and cone cells. The horizontal, bipolar, and amacrine cells combine the signals from various size regions, compare them to locate R6C-P.66
Images taken from taken Images interesting details, and pass that information on to higher levels in the visual cortex. 3/10/2020 DIGITAL IMAGE PROCESSING R6E24 -P.101 THE HUBEL AND WIESEL EXPERIMENTS
3/10/2020 DIGITAL IMAGE PROCESSING 25 CULTURAL DIFFERENCES ON WHAT WE SEE
Cultural differences strongly affect what we see in an image. Many westerners fix their attention on one (or a few) objects that are in the foreground and/or brightly colored, and ignore the surroundings. Many Asians pay more attention to the overall scene and the background details, noting the presence of objects in the foreground but not devoting any special attention to studying or remembering their characteristics. And, of course, recognition of something in a scene that appears familiar to the observer strongly influences where attention is focused. R6C-P.57 3/10/2020 DIGITAL IMAGE PROCESSING R6E26 -P.86-88 IS THERE ANY ANIMALS DOES NOT HAVE OR DOES NOT DEPEND ON VISUAL ORGAN?
Bats and dolphins use echolocation or sonar to probe the world about them. Pit vipers sense infrared radiation. Moles, living underground, trade sight for sensitive touch organs around their nose. Bloodhounds follow scents and butterflies have taste organs so sensitive they can detect single molecules. Some eels generate and sense electric fields to interact with their surroundings. Fish and alligators have pressure sensors that detect very slight motions in their watery environment.
R6C-P.55 3/10/2020 DIGITAL IMAGE PROCESSING 27 R6E-P.86 IMAGE FORMATION IN THE EYE
Flexible lens Controlled by the tension in the fibers of the ciliary body To focus on distant objects? To focus on objects near eye? Near-sighted and far-sighted
Light receptor Brain
radiant electrical energy impulses
3/10/2020 DIGITAL IMAGE PROCESSING IMAGE FORMATION IN THE EYE
2018) Muscles within the eye can be used to change the shape of the lens allowing us focus on objects that are near or far away An image is focused onto the retina causing rods and cones to become excited which ultimately send signals to the brain
x h
y f Figure Graphical representation of the eye looking at a This distance varies between 14-17 mm depending on the lens’ focussing palm tree. Point C is the focal center of the lens. 3/10/2020 DIGITAL IMAGE PROCESSING G3C-P.22 Images taken from Gonzalez & Woods, Digital Image Processing ( Processing Image Digital & Woods, fromGonzalez taken Images G3E-P.60-61 CLASSICAL OPTICAL THEORY
A ray passes through the centre C of the lens. x h The two triangle are proportional: = y f h is the height of the object on the retina (note that is located close to the fovea) Focal length of the eye: 17 to 14 mm Let h be the height in mm of that object in the retinal image, then 15/100 = h / 17 , h = 2.55mm The retinal image is reflected primarily in the area of the fovea.
3/10/2020 DIGITAL IMAGE PROCESSING 30 G3C-P.22 G3E-P.61 FOCUS
3/10/2020 DIGITAL IMAGE PROCESSING 31 FOCUS OF EYE
3/10/2020 DIGITAL IMAGE PROCESSING 32 FOCUS OF CAMERA
3/10/2020 DIGITAL IMAGE PROCESSING 33 PRESENTATION ASSIGMENT#3
为什么拍人像要用定焦镜头？什么定焦镜头拍人像比较好？
3/10/2020 DIGITAL IMAGE PROCESSING 34 PRESENTATION ASSIGMENT#4
相机和手机镜头的自动对焦的工作原理是什么
3/10/2020 DIGITAL IMAGE PROCESSING 35 BRIGHTNESS ADAPTATION & DISCRIMINATION
The range of light intensity human can adapt is enormous, roughly around 1010 to 1. Photopic vision alone has a range of around 106 to 1. The visual system does not operate simultaneously over the 1010 range. It accomplishes this large variation by changes in its overall sensitivity– brightness adaptation Similarly, the perceived intensity of a region is related to the light intensities of the regions surrounding it
3/10/2020 DIGITAL IMAGE PROCESSING G3C-P.22-23 G3E-P.61-62
BRIGHTNESS ADAPTATION 2018) Subjective brightness is a logarithmic function of the light intensity incident on the eye. Brightness adaptation: example “ Ba”
Figure Range of subjective brightness sensations showing a particular adaptation level,
mL = millilambert
3/10/2020 DIGITAL IMAGE PROCESSING 37 G3C-P.22 Images taken from Gonzalez & Woods, Digital Image Processing ( Processing Image Digital & Woods, fromGonzalez taken Images G3E-P.61-62
Vision
2017)
(
(3e) (3e)
CVIPtools
: Human and Computer Computer and Human :
Analysis
Applications with with Applications
Digital Image Processing and and Processing Image Digital
, , Umbaugh
Figure Brightness Adaptation in the Human Visual System (HVS). The entire range for the from Scott E E Scott from HVS is quite large, but only a small portion is used at a time based on current lighting levels; an example is shown in red. The pupil controls the amount of light allowed through, acting as
an automatic gain control (AGC), based on lighting conditions. Images taken taken Images 3/10/2020 DIGITAL IMAGE PROCESSINGUmbaugh3E38 -P.370 BRIGHTNESS ADAPTATION
Dynamic range of human visual system 10-6 ~ 104 Cannot accomplish this range simultaneously The current sensitivity level of the visual system is called the brightness adaptation level
3/10/2020 DIGITAL IMAGE PROCESSING 39 G3C-P.23 G3E-P.62 3/10/2020 DIGITAL IMAGE PROCESSING 40 3/10/2020 DIGITAL IMAGE PROCESSING 41 3/10/2020 DIGITAL IMAGE PROCESSING 42 BRIGHTNESS DISCRIMINATION
Brightness discrimination is the ability of the eye to discriminate between changes in light intensity at any specific adaptation level. Brightness discrimination is poor at low levels of illumination. The two branches in the curve indicate that at low levels of illumination vision is carried out by the rods, whereas at high level by the cones.
3/10/2020 DIGITAL IMAGE PROCESSING 43 G3C-P.23 G3E-P.62-63 BRIGHTNESS DISCRIMINATION
The quantity Ic/I, where Ic is the increment of illumination 2018) discriminable 50% of the time with background illumination I, is called the Weber ratio. A small value of Weber ratio, means good brightness discrimination. Larger Weber ratio indicates poor discrimination
Figure Basic experimental setup used to Figure A typical plot of the Weber ratio as a 3/10/2020 DIGITAL IMAGE PROCESSING 44 G3C-P.23 Images taken from Gonzalez & Woods, Digital Image Processing ( Processing Image Digital & Woods, fromGonzalez taken Images characterize brightness discrimination. function of intensity. G3E-P.62-63
2002)
( Inside(3e) Inside(3e)
Contrast sensitivity measurements
3/10/2020 DIGITAL IMAGE PROCESSING 45
Images taken from William K. Pratt , Digital Image Processing: PIKS PIKS Processing: Image K. Pratt, from William taken Digital Images Pratt3E-P.30 PERCEIVED BRIGHTNESS
Two phenomena clearly demonstrate that perceived brightness is not a simple function of intensity. - under/overshoot boundary of regions of different intensity (Mach bands) - A region’s perceived brightness does depend on the background intensity as well (simultaneous contrast)
Every light is a shade, compared to the higher lights, till you come to the sun; and every shade is a light, compared to the deeper shades, till you come to the night." —John Ruskin, 1879
3/10/2020 DIGITAL IMAGE PROCESSING G346 C-P.23-24 G3E-P.63 PSYCHOVISUAL EFFECTS
The perceived brightness is not a simple function of intensity Mach band pattern Simultaneous contrast And more… (see link)
3/10/2020 DIGITAL IMAGE PROCESSING 47 Images taken from Gonzalez & Woods, Digital Image Processing (2008) On the left side. In this image, edges between bars appear brighter on theright side and darker BRIGHTNESS ADAPTATION OF HUMAN HUMAN OF ADAPTATION BRIGHTNESS 3/10/2020 Intensities of surrounding Intensities of points point. effect perceived brightness at each intensities. regions ofdifferent of around boundary undershoot or overshoot Visual First Phenomena system tends to actual intensity. Perceived of simple functionnot a is intensity Figure Illustration of the Mach band effect. theMach of DIGITAL IMAGE PROCESSING EYE Umbaugh3E G3E G3 - P.63 C - P.376 - P.24 - 64
BRIGHTNESS ADAPTATION OF HUMAN EYE 2002)
( (CONT.) Inside(3e) Inside(3e)
A B
Ramp chart photo Intensity
Ramp chart intensity distribution Position
In area A, brightness perceived is darker while in area B is brighter. This phenomenon is called Mach Band Effect.
3/10/2020 DIGITAL IMAGE PROCESSING 49
Images taken from William K. Pratt , Digital Image Processing: PIKS PIKS Processing: Image K. Pratt, from William taken Digital Images Pratt3E-P.31 BRIGHTNESS ADAPTATION & DISCRIMINATION (CONT…)
An example of Mach bands
Mach E (1865) Über die Wirkung der raümlichen Verthelung des Lichtreizes auf die Netzhaut. Akad Wiss Lit Abh Math-Natwiss KL (Mainz) 52:303–322. Images taken from Gonzalez & Woods, Digital Image Processing (2002) Processing Image Digital Woods, & Gonzalez from taken Images Mach E (1886) The analysis of sensations and the relation of the physical to the psychical.3/10/2020 New York: Dover, 1959. DIGITAL IMAGE PROCESSING G3E-P.64
CRAIK-CORNSWEET-O’BRIEN STEP
2011)
(
Processing Handbook(6e) Handbook(6e) Processing
Image Image Russ, The The Russ, Craik KJW (1966) The nature of psychology (Sherwood SL, ed). (Cambridge UP, Cambridge, UK). Cornsweet TN (1970) Visual perception. (Academic, New York) O’Brien V (1959) Contrast by contour-enhancement. Am J Psychol 72:299–300. R6C-P.68 Images taken from taken Images 3/10/2020 DIGITAL IMAGE PROCESSING 51 R6E-P.104 BRIGHTNESS ADAPTATION OF HUMAN EYE (CONT.)
The second phenomena, called simultaneous contrast, a spot may appears to the eye to become darker as the background gets lighter.
An example of simultaneous contrast
Simultaneous contrast. All small squares have exactly the same intensity Images taken from Gonzalez & Woods, Digital Image Processing (2002) Processing Image Digital Woods, & Gonzalez from taken Images but they appear progressively darker as background becomes lighter.Umbaugh3E-P.377 3/10/2020 DIGITAL IMAGE PROCESSING 52 G3C-P.24 G3E-P.63-64
2011)
(
Processing Handbook(6e) Handbook(6e) Processing
Image Image Russ, The The Russ, Figure In the top row the lighter and darker surrounds affect our visual comparison of the (identical) central gray regions. This effect is reduced in the bottom row by the separating borders. R6C-P.71 Images taken from taken Images 3/10/2020 DIGITAL IMAGE PROCESSING 53 R6E-P.108
BRIGHTNESS ADAPTATION & DISCRIMINATION (CONT…)
2011)
(
Processing Handbook(6e) Handbook(6e) Processing
Image Image Russ, The The Russ,
DIGITAL IMAGE PROCESSINGUmbaugh3E-P.378 R6C-P.69 Images taken from taken Images 3/10/2020For more great illusion examples take a look at: http://web.mit.edu/persci/gaz/R6E54 -P.106 OPTICAL ILLUSION (VISUAL ILLUSION)
Under the influence of objective factors or under the control of their own psychological factors, the observers feel the wrong feelings that the graphics do not match the objective facts.
• Example: French tricolor
Blue: White: Red=30:33:37
White: expansion
Blue: contraction
3/10/2020 DIGITAL IMAGE PROCESSING 55 GREGORY’S CLASSIFICATION METHOD
• Three main classes: physical, physiological, and cognitive illusions
• Four kinds in each class
Ambiguities：(physical) an optical disturbance interveningberween the object and the retina
Distortions：(physical) disturbed physiological signals in the eyes or the brain
Paradoxes：(cognitive) application of misleading knowledge of objects
Fictions：(cognitive) application of misleading general rules Gregory, R. L. . (1997). Visual illusions classified. Trends in Cognitive Sciences,3/10/2020 1(5), 190-194. DIGITAL IMAGE PROCESSING 56 3/10/2020 DIGITAL IMAGE PROCESSING 57 PHYSICAL VISUAL ILLUSIONS
• Ambiguities：Mist
• Visibility in a natural haze declines exponentially with distance, due to scattering, which creates visual effects that are farther away.
3/10/2020 DIGITAL IMAGE PROCESSING 58 PHYSICAL VISUAL ILLUSIONS
• Distortions：Refraction
• Water has a refractive index of 1.33 and air has a refractive index of about 1. The bending of light rays as they move from the water to the air. Once the rays reach the eye, the eye traces them back as straight.
3/10/2020 DIGITAL IMAGE PROCESSING 59 PHYSICAL VISUAL ILLUSIONS
• Paradoxes：Mirror
• To be precise, it reverses the object in the direction perpendicular to the mirror surface (the normal).
3/10/2020 DIGITAL IMAGE PROCESSING 60 PHYSICAL VISUAL ILLUSIONS
• Fictions：Moiré pattern
• A visual result of interference between two lines or two objects at a constant angle and frequency.
• Will Discussed in L7？
3/10/2020 DIGITAL IMAGE PROCESSING 61 PHYSIOLOGICAL VISUAL ILLUSIONS
• Ambiguities：Distance of a single eye
• Because of the different positions of the eyes, the images seen are slightly different, which forms a three- dimensional image in the brain and makes people feel a sense of distance.
• Example：People can't close one eye and put the nibs of two pens together.
3/10/2020 DIGITAL IMAGE PROCESSING 62 PHYSIOLOGICAL VISUAL ILLUSIONS
• Distortions：Café wall[2]
• The illusion is only present when the mortar luminance lies between, or at least is not far outside, the luminances of the dark and light tiles.
• [3] explains the determinants of the tilt direction (Contrast polarities)
3/10/2020 DIGITAL IMAGE PROCESSING 63 PHYSIOLOGICAL VISUAL ILLUSIONS
• Paradoxes：Eye motion
• When visual channels disagree aftereffect of motion : moving yet not changing position or size.
• The visual system in the brain is too slow to process that information if the images are slipping across the retina at more than a few degrees per second.（e.g. Wagon-wheel effect）
• Will Discussed in L7？
3/10/2020 DIGITAL IMAGE PROCESSING 64 PHYSIOLOGICAL VISUAL ILLUSIONS
• Fictions：Afterimage
• A common physiological afterimage is the dim area that seems to float before one's eyes after briefly looking into a light source.
• Opponent color theory：red-green, blue-yellow, black-white.
3/10/2020 DIGITAL IMAGE PROCESSING 65 COGNITIVE VISUAL ILLUSIONS
• Ambiguities：Necker cube
• It is a simple wire-frame drawing of a cube with no visual cues as to its orientation, so it can be interpreted to have either the lower-left or the upper-right square as its front side.
3/10/2020 DIGITAL IMAGE PROCESSING 66 COGNITIVE VISUAL ILLUSIONS—— ILLUSIONS WITH TWO ALTERNATIVE INTERPRETATIONS
• Ambiguities：Rubin vase
• When the contours are not so unequal, ambiguity starts to creep into the previously simple inequality.
3/10/2020 DIGITAL IMAGE PROCESSING 67 COGNITIVE VISUAL ILLUSIONS—— ILLUSIONS WITH TWO ALTERNATIVE
INTERPRETATIONS
2011)
(
Processing Handbook(6e) Handbook(6e) Processing
Image Image Russ, The The Russ,
Do you see a young or an old lady? R6C-P. 73 Images taken from taken Images 3/10/2020 DIGITAL IMAGE PROCESSING 68 R6E-P.112 COGNITIVE VISUAL ILLUSIONS
• Distortions：Poggendorff
• Acute angles in the figure are seen by viewers as expanded.
3/10/2020 DIGITAL IMAGE PROCESSING 69 COGNITIVE VISUAL ILLUSIONS
Distortions
Orbison Vertical-horizontal illusion
3/10/2020 DIGITAL IMAGE PROCESSING 70 OPTICAL ILLUSIONS
Our visual systems play lots of interesting tricks on us
Optical illusions occurs when the eye fills in non-existing information or wrongly perceives geometrical properties of objects.
Müller-Lyer illusion Zöllner illusion
3/10/2020 DIGITAL IMAGE PROCESSING G3C-P.24 Images taken from Gonzalez & Woods, Digital Image Processing (2002) Processing Image Digital & Woods, fromGonzalez taken Images G3E-P.64-65
Vision
2017)
(
(3e) (3e)
CVIPtools
: Human and Computer Computer and Human :
Analysis
Applications with with Applications
Digital Image Processing and and Processing Image Digital
, , Umbaugh
Figure Optical Illusions. (a) Even though the vertical lines are parallel, they appear
tilted, (b) The top line appears shorter than the bottom one,(c) the two diagonal line from Scott E E Scott from segments appear not to be collinear, (d) is this nine black crosses, or connected white rectangles? (e) the outer two lines in the upper group appear to be farther apart than the two lines in the lower group, and (f) the two center circles are the same size, but the Images taken taken Images 3/10/2020 DIGITAL IMAGE PROCESSING 72 one surrounded by larger circles appears smaller. Umbaugh3E-P.378 EBBINGHAUS ILLUSION
3/10/2020 DIGITAL IMAGE PROCESSING 73
THE “TOP HAT” ILLUSION
2011)
(
Processing Handbook(6e) Handbook(6e) Processing Image Image
Russ, The The Russ, In this exaggerated drawing of a top hat, the height of the crown appears to be much greater than the width of the brim, but in fact they are exactly the same.
R6C-P.86 Images taken from taken Images 3/10/2020 DIGITAL IMAGE PROCESSING 74 R6E-P.131 MORE …
Do you see squares?
More at http://scientificpsychic.com/graphics/index.html 3/10/2020 DIGITAL IMAGE PROCESSING 75
FRASER’S SPIRAL
2011) (
The circular rings are visually “tilted”
Processing Handbook(6e) Handbook(6e) Processing and perceived to
form a spiral
Image Image Russ, The The Russ,
Fraser J (1908) A New Visual Illusion of Direction. British Journal of Psychology 2:307–320 R6C-P.72 Images taken from taken Images 3/10/2020 DIGITAL IMAGE PROCESSING 76 R6E-P.110 COGNITIVE VISUAL ILLUSIONS
• Distortions：Ames room
• An Ames room is viewed with one eye through a peephole. Through the peephole the room appears to be an ordinary rectangular cuboid.
3/10/2020 DIGITAL IMAGE PROCESSING 77 3/10/2020 DIGITAL IMAGE PROCESSING 78 COGNITIVE VISUAL ILLUSIONS
• Paradoxes：Penrose impossible objects
• It consists of a two-dimensional figure which is instantly and subconsciously interpreted by the visual system as representing a projection of a three-dimensional object.
3/10/2020 DIGITAL IMAGE PROCESSING 79
2011)
(
Processing Handbook(6e) Handbook(6e) Processing
Image Image Russ, The The Russ,
R6C-P. 74 Images taken from taken Images 3/10/2020 DIGITAL IMAGE PROCESSING 80 R6E-P.113 COGNITIVE VISUAL ILLUSIONS
• Fictions：Illusory contours
• Objects in the natural world are often only partially visible. Illusory contours provide clues for how the visual system constructs surfaces when portions of the surface’s edge are not visible.
3/10/2020 DIGITAL IMAGE PROCESSING 81
2011)
(
Processing Handbook(6e) Handbook(6e) Processing
Image Image Russ, The The Russ,
Figure Visually connecting each point to its nearest neighbor produces the impression of radial or circumferential alignment. R6C-P.70 Images taken from taken Images 3/10/2020 DIGITAL IMAGE PROCESSING 82 R6E-P.108 EXPLANATION PERCEPTUAL ORGANIZATION
• To make sense of the world it is necessary to organize incoming sensations into information which is meaningful by perceiving individual sensory stimuli as a meaningful whole.
• Continuity：the perceptual system tries to disambiguate which segments fit together into continuous lines
• Similarity：objects that are similar are seen as associated
3/10/2020 DIGITAL IMAGE PROCESSING 83 EXPLANATION DEPTH AND MOTION PERCEPTION
• Depth Perception：Ponzo illusion
• The converging parallel lines tell the brain that the image higher in the visual field is farther away, therefore, the brain perceives the image to be larger.
• Motion Perception：Phi phenomenon
• Film animation is based on the illusion that the brain perceives a series of slightly varied images produced in rapid succession as a moving picture.
3/10/2020 DIGITAL IMAGE PROCESSING 84 EXPLANATION COLOR AND BRIGHTNESS CONSTANCIES
• An illusion of color difference or luminosity difference can be created when the luminosity or color of the area surrounding an unfamiliar object is changed.
• The luminosity of the object will appear brighter against a black compared to a white field.
3/10/2020 DIGITAL IMAGE PROCESSING 85
2011)
(
Processing Handbook(6e) Handbook(6e) Processing
Image Image Russ, The The Russ, Color interpretation is influenced by grouping with surroundings. The red colors marked in the two spirals are actually identical, but are perceived as magenta and orange because of the adjacent blue and yellow stripes. (Image courtesy of Gregory Francis, Purdue Univ.) R6C-P.71 Images taken from taken Images 3/10/2020 DIGITAL IMAGE PROCESSING 86 R6E-P.109 3/10/2020 DIGITAL IMAGE PROCESSING 87
2011)
(
Processing Handbook(6e) Handbook(6e) Processing
Image Image Russ, The The Russ,
The regions marked by arrows are identical white, but the colors from the adjacent lines are visually extended into the spaces and are perceived as adding color to the regions. (Image courtesy of Gregory Francis, Purdue Univ.) R6C-P.71 Images taken from taken Images 3/10/2020 DIGITAL IMAGE PROCESSING 88 R6E-P.109 EXPLANATION OBJECT
• The brain has the ability to understand familiar objects as having a consistent shape or size.
• Unfamiliar objects：shape may change when the perspective is changed.
3/10/2020 DIGITAL IMAGE PROCESSING 89 EXPLANATION COGNITIVE PROCESSES HYPOTHESIS
The hypothesis claims that visual illusions occur because the neural circuitry in our visual system evolves, by neural learning, to a system that makes very efficient interpretations of usual 3D scenes based in the emergence of simplified models in our brain that speed up the interpretation process but give rise to optical illusions in unusual situations.
3/10/2020 DIGITAL IMAGE PROCESSING 90
2011)
(
Processing Handbook(6e) Handbook(6e) Processing
Image Image Russ, The The Russ, Rotating the same image (cuneiform indentations in a clay tablet) by 180 degrees makes the pits (at left) appear to be peaks (at right). R6C-P.84 Images taken from taken Images 3/10/2020 DIGITAL IMAGE PROCESSING 91 R6E-P.129 3/10/2020 DIGITAL IMAGE PROCESSING 92 OPTICAL ILLUSIONS (CONT…) Stare at the cross in the middle of the image and think circles
3/10/2020 DIGITAL IMAGE PROCESSING 93 BENHAM'S DISK
3/10/2020 DIGITAL IMAGE PROCESSING 94 BLUE-BLACK OR WHITE-GOLD? WHAT COLOR IS THE DAMN DRESS!
3/10/2020 DIGITAL IMAGE PROCESSING 95 JUST SOME FUN VISUAL PERCEPTION GAMES
Can you count the dots?
3/10/2020 DIGITAL IMAGE PROCESSING 96 MIND MAP EXERCISE: MIND MAPPING FOR NOTE TAKING
Beau Lotto: Optical Illusions Show How We See http://www.ted.com/talks/lang/eng/beau_lotto_optical_illusions_show_how_we_see.html3/10/2020 DIGITAL IMAGE PROCESSING 97 APPLICATION IN ART
Artists who have worked with optical illusions：M. C. Escher
3/10/2020 DIGITAL IMAGE PROCESSING 98 APPLICATION IN ART
Contemporary artists who have experimented with illusions：Patrick Hughes
3/10/2020 DIGITAL IMAGE PROCESSING 99 APPLICATION OP ART
Short for optical art, is a style of visual art that uses optical illusions.
3/10/2020 DIGITAL IMAGE PROCESSING 100 APPLICATION TROMPE-L‘ŒIL
Uses realistic imagery to create the optical illusion that the depicted objects exist in three dimensions.
3/10/2020 DIGITAL IMAGE PROCESSING 101 APPLICATION FORCED PERSPECTIVE
• A technique which employs optical illusion to make an object appear farther away, closer, larger or smaller than it actually is.
• It has applications in photography, filmmaking and architecture.
An 8.6-metre-long gallery 3/10/2020 DIGITAL IMAGE PROCESSING 102 APPLICATION FORCED PERSPECTIVE
• Using the monocular cues （erial perspective, blurring, relative size and lighting） in concert with angular size, the eyes can perceive the distance of an object.
• The subtended angle increases as the object moves closer to the lens. Two objects with different actual size have the same apparent size when they subtend the same angle.
• Formula for calculating angular size：
3/10/2020 DIGITAL IMAGE PROCESSING 103 FURTHER READINGS
Gregory, R. L. . (1997). Visual illusions classified. Trends in Cognitive Sciences, 1(5), 190-194. Gregory, R. L. , & Heard, P. . (1979). Border locking and the café wall illusion. Perception, 8(4), 365-380. Kitaoka, A. , Pinna, B. , & Brelstaff, G. . (2004). Contrast polarities determine the direction of café wall tilts. Perception, 33(1), 11-20. http://www.archimedes-lab.org/index_optical.html http://www.moillusions.com/ https://michaelbach.de/ot/
3/10/2020 DIGITAL IMAGE PROCESSING 104 FURTHER READINGS
John P. Frisby (1980) Illusion, Brain and Mind, Oxford Univ. Press John P. Frisby and James V. Stone (2010) Seeing, 2nd Edition: The Computational Approach to Biological Vision, MIT Press, Boston David Marr (1982) Vision, W. H. Freeman Co. Irvin Rock (1984) Perception, W. H. Freeman Co. Steven H. Schwartz Visual Perception McGraw-Hill The Joy of Visual Perception http://www.yorku.ca/eye/thejoy.htm
3/10/2020 DIGITAL IMAGE PROCESSING 105 THE END OF LECTURE 2
3/10/2020 DIGITAL IMAGE PROCESSING 106