UNIT – 7

•Color Models: RGB, CMY, HSV,

•Color and light,
•color and objects-
•color and the eye-
•color consistency-
•color terms
•reactions to color
•and color continuation-
•color on engineering equipments. 06 Hours
Color models
• Color models
 RGB
 YCbCr (YPbPr)
 YUV
 YIQ
 CMYK
• A comparison of them

3
 Color Space
• What is color space?
 A 3D model used to define a specified color
• The difference between color spaces:
 The choice of axes

4
 Color Space – RGB

• RGB:
 The simplest color space
 Axes: Red, green, blue
 Advantages: simple

5
 Color Space – YCbCr &YPbPr
• YCbCr & YPbPr
 Used for: digital video encoding, digital
camera
• Axes:
 Y: luma
 Cb: blue chroma
 Cr: red chroma

6
 Color Space – YCbCr &YPbPr
• Conversion from RGB:
 Y=0.299(R-G) + G + 0.114(B-G)
 Cb=0.564(B-Y)
 Cr=0.713(R-Y)
• The Matrix form:

 Y   0.299 0.587 0.114  R 

    
Cb
     0.168636 0.232932 0.064296 G
 
 Cr   0.499813 0.418531 0.081282  B 
    
7
 Color Space – YCbCr &YPbPr
• Why do we use the luma & chroma channel?

• Advantage:
 Bandwidth efficiency

8
 Color Space – YUV
• YUV
 Used for: video encoding for some standard
such as NTSC, PAL, SECAM
• Axes:
 Y: luma
 U: blue chroma
 V: red chroma

9
 Color Space – YUV
• Conversion from RGB:
 Y=0.299R+0.587G+0.114B
 U=0.436(B-Y)/(1-0.114)
 V=0.615(R-Y)/(1-0.299)
• The Matrix form:

 Y   0.299 0.587 0.114  R 

    
 U    0.14713 0.28886 0.436  G 
 V   0.615 0.51499 0.10001 B 
    
10
 Color Space – YIQ
• YIQ
 Used for: video encoding for some standard such as
NTSC
• Axes:
 Y: luma
 I: blue chroma
 Q: red chroma
• I-Q channels are rotated from the U-V channels in
YUV

11
 Color Space – YIQ
• Conversion from RGB:

 Y   0.299 0.587 0.114  R 

    
I
   0.595716 0.274453  0.321263 G
 
 Q   0.211456 0.522591 0.311135  B 
    

12
 Color Space – CMYK
• Used for: printer printing
• Use the subtractive color mixing
• Axes:
Cyan
Magenta
Yellow
K: black

13
 Color Space – CMYK
• Conversion from RGB:

 C = 255 -Y - 1.4021(Cr-128)
 M = 255 - Y + 0.3441(Cb-128) + 0.7142(Cr-128)
 Y = 255 - Y - 1.7718(Cb -128)
 K = min (C, M, Y)

14
 Color Space – Comparison

Color Color Primary Used for Pros and

space mixing parameters cons
RGB Additive Red, Easy but wasting
Green, Blue bandwidth

CMYK Subtractive Cyan, Magenta, Printer Works in pigment

Yellow, Black mixing
YCbCr additive Y(luminance), Video encoding, Bandwidth efficient
YPbPr Cb(blue chroma), digital camera
Cr(red chroma)
YUV additive Y(luminance), Video encoding Bandwidth efficient
U(blue chroma), for NTSC, PAL,
V(red chroma) SECAM

YIQ additive Y(luminance), Video encoding Bandwidth efficient

I(rotated from U), for NTSC
Q(rotated from V)
15
 HSV: Hue, Saturation and Value
• Hue: spectral hues + mixed colours like purple
• Saturation: purity of colour
• Value: brightness or intensity

purple!

2002/02/05 PSYC202-005, Term 2, Copyright Jas 16

on Harrison, 2002
 HSV: Hue, Saturation and Value
Hue Saturation Value

high high

low low

2002/02/05 PSYC202-005, Term 2, Copyright Jas 17

on Harrison, 2002
 HSV: a psychological colour space

hue

2002/02/05 PSYC202-005, Term 2, Copyright Jas 18

on Harrison, 2002
 HSV: a psychological colour space

saturation

2002/02/05 PSYC202-005, Term 2, Copyright Jas 19

on Harrison, 2002
 HSV: a psychological colour space
brightness
high

medium

low

2002/02/05 PSYC202-005, Term 2, Copyright Jas 20

on Harrison, 2002
 HSV: a psychological colour
white space

2002/02/05 PSYC202-005, Term 2, Copyright Jas black 21

on Harrison, 2002
Colour Terms
Colours
•Colours are one of the important sensation factor in industrial design and
ergonomics

•Colours describes those sensation produced in brain as rays of differing

wavelength impinge upon eyes retina

•Light is necessary to detect colours

• Eye is sensitive to visual light which lies between 380 to 760 nanometer
Electromagnetic Spectrum
Colour terms

Colour terms are the System which is specify to understand any specific
colour.

Colour terms are useful to identify two types of colours.

•Chromatic colour
•Achromatic colour

Chromatic colour - any colour which lies between the spectrum

i e – VIBGOYR

Achromatic colour – colour that is white through all shades of grey to black
History of Identifying Colors

Newton – discovered spectrum and devised color circle

The system for colour identification was established in 1931 by
the commission of international eclairage (CIE)

Later others came up with thier colour notation systems (colour

circles)

•Newton colour circle

•Maxwell colour triangle
•Ives colour circle
•Ostwald Colour circles
•Munsell Colour System
Colour circles – Newtons colour circle
Maxwell colour triangle
Ives colour system
 Ostwald's color wheel

•Ostwald’s double-cone:
• Ostwald's color wheel or double cone in its simplest diagrammatic form. The
colors are at their most saturated on the equator; from this point they become
lighter toward white or darker toward black in the direction of the poles of the
neutral axis. The inside of the solid contains all the colors it is possible to mix with
their relative gradations of lightness and darkness.
Munsell Colour
System

•Hue

•chroma
(describing
satuartion)

•value (describing
lightness)
chroma

Value

Hue
Munsell systems are interpreted as follows-

• Hue
• Munsell value (describing lightness)
• chroma (describing satuartion)

Hue - Shows all variations of colors/hues, which distinguishes one

chromatic colour from another on the munsell system

Notation: 5.0 + letter

5.0 R = red
5.0 RP = red purple
Value Scale

• Vertical arrangement: the “trunk” of the color tree

• The lightness-darkness of a particular hue
• Related to brightness

Nine levels of values

1 = black
9 = white

Eg
Notation: 5.0 R +5/ ; 5.0+ BG 7/
• Chroma: (describing saturation) – Describes the purity of colour
Eg
notation of color is 5+RP 5/26 indicates a red purple hue.
5+RP(Hue) ,value 5 and chroma 26.
Other colour terms

One of the difficulty with munsell system is that the expression Hue, value,
chroma may not familiar to users

•Hue is equivalent to colour

•Saturation is equivalent to chroma
•Tone is equivalent to value
&
•Tint is equivalent to any colour of high value
•Shade is equivalent to any colour of low value
 PSYCHOLOGICAL EFFECTS
• Red
 PSYCHOLOGICAL EFFECTS
BLUE
 PSYCHOLOGICAL EFFECTS
• YELLOW
 COLOUR AND MACHINE FORM:

 Colour has a strong influence

upon the apparent form of an
object. Dark colours giving
the effect of weight may be
used on foundation or main
supporting structures in order
to underline their character,
while light colours may be
used to minimise the effect of
masses which might otherwise
create a top-heavy effect.
Which of these boxes do you think is bigger/heavier?
 In many cases a two-colour or three-colour
treatment can produce all that is needed.
Variations need to be made by changing hue.
• Despite the different ways in which colour can
be applied ,the number of colours used should
be kept to a minimum on the practical
grounds of cost. Too many colours are likely to
make a machinery appear garish.
 COLOUR AND STYLE

Colour styles in industrial equipment depend

upon choosing colours which identify the
manufacturer.
• Use of chromium plating and high gloss
finishes to create an impression of fine
workmanship is unnecessary.
• The result is usually lowering in apparent
worth while creating distraction to the
machine operator.
 CONCLUSION
• Colour should be regarded as a tool to help
operator first then, as means for other aspects.
• Any intellectual approach will not lead to
particular colours but it will determine ranges
of colours from which particular colours can
be chosen .
 Recovery of World Structure:
Art and Image

PSYC202, Term 2, Copyright Jason Harrison

2002/02/12 62
2002
 Task of visual perception
• Stay alive, reproduce, make art
• Estimate properties of world
using light falling on retina
• Information:
– originates from light reflected from objects
– internal knowledge (evolutionary/experience)
• Combined to recover “true” object properties

2002/02/12 PSYC202, Term 2, Copyright Jason H 63

arrison 2002
 Complications, complications…
1. Illumination (intensity, hue, motion)
2. Projection (3D world to 2D images)
3. Fragmentation (occlusion)

2002/02/12 PSYC202, Term 2, Copyright Jason H 64

arrison 2002
 1. Illumination
• intensity, hue, motion
• varies from spot to spot, moment to moment
• recover “true” surface properties

What material is this?

2002/02/12 From http://www.discount-line.com/bath.htm

PSYC202, Term 2, Copyright Jason H 65
arrison 2002
 Projection (3D world to 2D images)
• recover depth and spatial organization
• depth enables two abilities:
a. shape constancy
true 3D shape of object
b. size constancy
true 3D size of object

2002/02/12 PSYC202, Term 2, Copyright Jason H 66

arrison 2002
 Fragmentation (occlusion)
• objects not always connected in image

2002/02/12 FromPSYC202,
http://www.heavenforcars.com/products/gifts/index7.htm
Term 2, Copyright Jason H 67
arrison 2002
 Complications, complications…
1. Illumination (intensity, hue, motion)
2. Projection (3D world to 2D images)
3. Fragmentation (occlusion)

2002/02/12 PSYC202, Term 2, Copyright Jason H 68

arrison 2002
 Projection
a. shape constancy
• Perceived shape
of object does
not change as
orientation
changes
• Shape of image
on retina
changes

2002/02/12 From http://zeus.rutgers.edu/~ikovacs/SandP/prepIII_1.html

PSYC202, Term 2, Copyright Jason H 69
arrison 2002
 Projection
a. shape constancy
Same image could be due
to:
- different shapes
- different orientations
- different depths

2002/02/12 PSYC202, Term 2, Copyright Jason H 70

arrison 2002
 Projection
a. shape constancy
• if shape known
– depth and
orientation can be
“guessed”

2002/02/12 PSYC202, Term 2, Copyright Jason H 71

arrison 2002
 Projection
a. shape constancy
• if depth known
– shape and
orientation can
be “guessed”

distance 1

distance 2

2002/02/12 PSYC202, Term 2, Copyright Jason H 72

arrison 2002
 Projection
b. size constancy
• known object size
• retinal size determines
depth
• depth cues determine size

2002/02/12From http://psychlab1.hanover.edu/classes/Sensation/SizeConstancy/index.html
PSYC202, Term 2, Copyright Jason H 73
arrison 2002
 Projection
b. size constancy
• known object size
• retinal size determines
depth
• depth cues determine size

2002/02/12 PSYC202, Term 2, Copyright Jason H 74

arrison 2002
 Projection
b. size constancy
• known object size
• retinal size determines
depth
• depth cues determine size

2002/02/12 PSYC202, Term 2, Copyright Jason H 75

arrison 2002
 Projection
b. size constancy

pupil

retinal
image

2002/02/12 PSYC202, Term 2, Copyright Jason H 76

arrison 2002
 Projection
b. size constancy

pupil

retinal
image s
(size = r)

distance d

2002/02/12 PSYC202, Term 2, Copyright Jason H 77

arrison 2002
 Projection
b. size constancy

pupil

retinal
image  s
(size = r)
visual
angle 

distance d

2002/02/12 PSYC202, Term 2, Copyright Jason H 78

arrison 2002
 Projection
b. size constancy

pupil

retinal
image  s
(size = r)
visual
angle 

tan s/d
or
arctan(s/d)

distance d

-angle q depends only on ratio s/d

2002/02/12 PSYC202, Term 2, Copyright Jason H 79
arrison 2002
 Projection
b. size constancy

pupil

retinal
image  s1
(size = r)
visual
angle 

tan s/d
or
arctan(s/d)

distance d1

2002/02/12 PSYC202, Term 2, Copyright Jason H 80

arrison 2002
 Projection
b. size constancy
• Angle q the same
as long as s/d the
pupil
same
retinal
• larger or smaller image  s1 s2
(size = r)
objects can have visual
angle 
same q
- 1 = s2 tan s/d
or
d1 d2 arctan(s/d)

distance d1

distance d2

2002/02/12 PSYC202, Term 2, Copyright Jason H 81

arrison 2002
 • Projection
b. size constancy
• q is known
(from retinal pupil

size) -> tan

q retinal
image  s1
(size = r)
• if size is known, visual
angle 
distance follows
• if distance is tan s/d
or
known, size arctan(s/d)

follows distance d1

2002/02/12 PSYC202, Term 2, Copyright Jason H 82

arrison 2002
 Size and Shape Constancy
• depend on experience
• depend on expectations
• “top down” processes

• how else can we recover depth from images?

• what depth cues are available?

2002/02/12 PSYC202, Term 2, Copyright Jason H 83

arrison 2002
 Depth cues
1. Pictorial cues (single image)
– art techniques (perspective, shading, occlusion)

2. Multi-point cues (multiple images)

– motion cues (kinetic depth, motion parallax)
– binocular cues

3. Physiological cues (real stimuli)

– accommodation of lens
– vergence movements
2002/02/12 PSYC202, Term 2, Copyright Jason H 84
arrison 2002
 Pictorial cues
• occlusion/interposition
• shading
• shadows
• perspective
• texture gradient
• height in plan
• aerial perspective
• familiar size

2002/02/12 PSYC202, Term 2, Copyright Jason H 85

arrison 2002
 Occlusion/Interposition
• nearer object occludes further one

2002/02/12 PSYC202, Term 2, Copyright Jason H 86

arrison 2002
 Occlusion/Interposition
• nearer object occludes further one
• critical feature “T-junction”

crossbar belongs to
object in front

T-junction

2002/02/12 PSYC202, Term 2, Copyright Jason H 87

arrison 2002
 Shading
• lightness constancy: shading determines
shape
• uniform reflectance: 3D shape from shading
• lighting from above
– bumps are light on top
– dents are light on bottom

2002/02/12 PSYC202, Term 2, Copyright Jason H 88

arrison 2002
 Countershading
confuse interpretation of shape

2002/02/12 PSYC202, Term 2, Copyright Jason H 89

arrison 2002
 Countershading on the
BB-55 USS North Carolina (1937-1961)
• more difficult to see or determine barrel diameter

2002/02/12 PSYC202,From
Term 2,http://www.navsource.org/archives/01/55.htm
Copyright Jason H 90
arrison 2002
 Shadows
• dark patches interpreted as shadows
• lighting from above
with shadows perceived as
standing (3D)

Ball and shadow demo

2002/02/12 PSYC202, Term 2, Copyright Jason H 91
arrison 2002
 Shadows
• dark patches interpreted as shadows
• lighting from above
without shadows perceived as
flat (2D)

(applications to computer interfaces)

Ball and shadow demo

2002/02/12 PSYC202, Term 2, Copyright Jason H 92
arrison 2002
 Perspective
• parallel lines in world converge to a vanishing
point in image
• lines that converge are parallel in world
VP 1 VP 2

2002/02/12 PSYC202, Term 2, Copyright Jason H 93

arrison 2002
 Texture gradient
• perspective: location of texture items and
• changes in size/shape of texture items

2002/02/12 PSYC202, Term 2, Copyright Jason H 94

arrison 2002
 Change in texture gradient:
change in shape and depth

From http://psych.hanover.edu/Krantz/art/texture.html
2002/02/12 PSYC202, Term 2, Copyright Jason H 95
arrison 2002
 Height in plane (distance from horizon)
• objects closer to horizon appear further away

Same object, one closer to horizon: smaller or further away?

2002/02/12 PSYC202, Term 2, Copyright Jason H 96
arrison 2002
 Aerial perspective (da Vinci)
• smog, haze, etc. scatter light
– objects father away: more scattering
– if object surrounded by haze, appears farther
away

2002/02/12 PSYC202, Term 2, Copyright Jason H 97

arrison 2002
 Aerial perspective (da Vinci)
• smog, haze, etc. scatter light
– objects father away: more scattering
– if object surrounded by haze, appears farther
away Without haze, objects
appear equally distant

2002/02/12 PSYC202, Term 2, Copyright Jason H 98

arrison 2002
 Aerial perspective
• Why do mountains look closer on a clear day?
“Whoa, look at the mountains!”
• Visual system assumes average amount of
haze
– more haze, more distance object must be
– less haze, less distant object must be

– on a clear day, less haze than usual

• mountains appear closer than normal

2002/02/12 PSYC202, Term 2, Copyright Jason H 99

arrison 2002
 Familiar size
• if size of object is known
– perceived size = true size
– depth is inferred

2002/02/12 PSYC202, Term 2, Copyright Jason H 100

arrison 2002
 1. Pictorial cues
• occlusion/interposition
• shading
• shadows
• perspective
• texture gradient
• height in plan
• aerial perspective
• familiar size

2002/02/12 PSYC202, Term 2, Copyright Jason H 101

arrison 2002
 2. Multi-point cues
• kinetic depth (motion of object)
• motion parallax (motion of observer)
• binocular stereopsis

• (see lecture from January 29 for kinetic depth

and motion parallax)

2002/02/12 PSYC202, Term 2, Copyright Jason H 102

arrison 2002
 Binocular Stereopsis
• recovery of 3D depth via 3D displacement of eyes
• each eye sees slightly different image
Position 1 (Overhead view) Position 2 (Overhead view)

Observer sees Observer sees

cube to the cube to the
rig ht o f s p he re : le ft o f s p he re :

2002/02/12 PSYC202, Term 2, Copyright Jason H 103

arrison 2002
 Binocular Stereopsis
• recovery of 3D depth via 3D displacement of eyes
• each eye sees slightly different image
Position 1 (Overhead view) Position 2 (Overhead view)

Left eye sees Right eye sees

cube to the cube to the
rig ht o f s p he re : le ft o f s p he re :

Left eye Right eye

2002/02/12 PSYC202, Term 2, Copyright Jason H 104
arrison 2002
 Binocular
• amount of difference Stereopsis
between image depends on depth
– depth can be calculated exactly from differences
– as if angles between eyes were used
• limitation: depth less than 3 meters
• limitation: requires matching images (escalator)

2002/02/12 PSYC202, Term 2, Copyright Jason H 105

arrison 2002
 Binocular Stereopsis: stereoscope
• use two static images one for each eye
– coloured glasses, polarized lenses, lenses
– divergence of eyes: practice, practice, practice

Image for left eye

Image for right eye

2002/02/12 PSYC202, Term 2, Copyright Jason H 106

arrison 2002
 Binocular Stereopsis
• there are always two slightly different images
entering visual system
• double vision rarely occurs
• to avoid double vision (binocular rivalry)
– suppress input from one eye
– dominant eye is “attended”
– non-dominant eye is “unattended”
– if two completely different images are presented
• attention (and cortex activity) switches back and forth
2002/02/12 PSYC202, Term 2, Copyright Jason H 107
arrison 2002
 Binocular rivalry
Tong et al 1998
• two different images
– face and house
• presented either
– simultaneously
– sequentially
• same fMRI responses
• rivalry resolved below high-
level areas
From http://www.princeton.edu/~ftong/
2002/02/12 PSYC202, Term 2, Copyright Jason H 108
arrison 2002
 2. Multi-point cues
• kinetic depth (motion of object)
• motion parallax (motion of observer)
• binocular stereopsis

• (see lecture from January 29 for kinetic depth

and motion parallax)

2002/02/12 PSYC202, Term 2, Copyright Jason H 109

arrison 2002
 3. Physiological cues - real stimuli
• accommodation
– focusing of an eye
• vergence
– angle between two eyes

2002/02/12 PSYC202, Term 2, Copyright Jason H 110

arrison 2002
 Accommodation
• lens changes (one eye)
shape to focus on objects
– rounder for near object
– flatter for far objects
• limitation: depth between 20cm and 2m

2002/02/12 PSYC202, Term 2, Copyright Jason H 111

arrison 2002
 Vergence (two eyes)
Eyes fixate on object - rotate inwards if near by

Eyes fixate on object - less rotation if far away

2002/02/12 PSYC202, Term 2, Copyright Jason H 112

arrison 2002
 Vergence
• difference from binocular stereopsis
– angle of eyes rather than different images
• limitation: depth less than 3m

2002/02/12 PSYC202, Term 2, Copyright Jason H 113

arrison 2002
 Complications, complications…
1. Illumination (intensity, hue, motion)
2. Projection (3D world to 2D images)
3. Fragmentation (occlusion)

2002/02/12 PSYC202, Term 2, Copyright Jason H 114

arrison 2002
 Fragmentation
• objects connected in world
– not always connected in image on retinas
• determine which patches in image belong
together
• Visual completion
B
A

2002/02/12 PSYC202, Term 2, Copyright Jason H 115

arrison 2002
 Visual completion (fragmentation)
• Two abilities
– modal completion: completion of pieces
fragmented by occluders in world
– blindspot completion: completion of visual image
fragmented by existence of blindspot

2002/02/12 PSYC202, Term 2, Copyright Jason H 116

arrison 2002
 Modal completion
• handle fragmentation due to external occluder

2002/02/12 PSYC202, Term 2, Copyright Jason H 117

arrison 2002
 Modal completion
• handle fragmentation due to external occluder

2002/02/12 PSYC202, Term 2, Copyright Jason H 118

arrison 2002
 Modal completion
• handle fragmentation due to external occluder

120°

2002/02/12 PSYC202, Term 2, Copyright Jason H 119

arrison 2002
 Modal completion
• handle fragmentation due to external occluder

2002/02/12 PSYC202, Term 2, Copyright Jason H 120

arrison 2002
 Modal completion
• handles fragmentation due to external occluder

• Result: - pieces are linked when

- edges intersect at >= 90°
- there is a visible occluder
• but no “filling in” behind occluder
2002/02/12 PSYC202, Term 2, Copyright Jason H 121
arrison 2002
 Blindspotdue
• handles fragmentation completion
to blindspot in eye

"Hole"
where
optic
nerve
leaves
retina
= blindspot Optic
Retina nerve

2002/02/12 PSYC202, Term 2, Copyright Jason H 122

arrison 2002
 Blindspot completion
Pattern of light
on retina
Blind spot

Information actually
picked up

?
What kind of completion
occurs?
2002/02/12 PSYC202, Term 2, Copyright Jason H 123
arrison 2002
• Result - areaBlindspot completion
is filled in (limited)
• filling in includes
– texture, complex patterns (but not text)
– smooth transitions between colours, alignment,
intensity, etc.

123456
AB
891012
345678 CD

2002/02/12 PSYC202, Term 2, Copyright Jason H 124

arrison 2002
 Recovery of World Structure:
Art and Image

1.Illumination (intensity, hue, motion)

2.Projection (3D world to 2D images)
3.Fragmentation (occlusion)

2002/02/12 PSYC202, Term 2, Copyright Jason H 125

arrison 2002