Digital Image Processing

Dr. M. Ilyas Fakhir

Lecture-12
 2
of
36
Color Image Processing
 The use of color in image processing is motivated
by:
 Color is important in object recognition
 Human eyes can discern thousands of colors
 Color image processing:
 Full color image processing
 Pseudo color processing
 Full color processing: image is acquired by a full-
color sensor
 Pseudo color processing: assigning a shade of
color to a particular monochrome intensity or
range of intensities
 3
of
36
Color Image Processing
 In 1666 Newton discovered that a beam of sunlight
passed through a prism will break into a spectrum
of colors ranging from violet at one end to red at the
other
 Color spectrum: violet, blue, green yellow, orange,
and red.
 No color in the spectrum end abruptly
 4
of
36
Color Image Processing
 Color perceived from an object is determined by the
nature of light reflected from that object.
 An object reflecting light that is balanced in all
visible light appears white.
 An objects that favors reflectance in a limited range
of the visible spectrum exhibits a specific color
 5
of
36
Color Image Processing
 Achromatic light (without color) is described by
intensity (amount)
 Chromatic light is described by 3 quantities:
 Radiance
 Luminance
 Brightness
 Radiance: total amount of energy that flows from
the light source
 Luminance: a measure of the amount of energy
an observer perceives from a light source
 Infrared source: zero luminance
 Brightness: a subjective quantity
 6
of
36
Color Image Processing
 Cones can be divided into three principal sensing
categories: red, green and blue
 7
of
36
Color Image Processing
 Structure of human eye: all colors are seen as
variable combinations of 3 primary colors, Red,
Green and Blue
 Primary colors can be added to produce secondary
colors:
 Magenta: (red plus blue)
 Cyan: (green plus blue)
 Yellow: (red plus green)
 Mixing three primary colors produce white.
 Mixing a secondary color with its opposite primary
color produces white
 8
of
36
Color Image Processing
 Color TV: an example of the additive nature of light
colors
 Interior of TV tube: a large array of triangular dot
patterns of electron sensitive phosphor
 Each dot in a triangle produces one of the primary
colors
 Intensity of red-emitting phosphor is modulated by
an electron gun.
 Similarly for green-emitting and blue-emitting
 Three primary colors are added and received by
the eye as a full-color image
 9
of
36
Color Image Processing
 One color from another is distinguished by 3
factors:
1. Brightness
2. Hue
3. Saturation
 Hue: dominant wavelength (color) in a mixture of
light waves
 Saturation: relative purity or the amount of white
light mixed with a hue
 Pink is less saturated than red
10
of
36
Color Image Processing
 Pigment =
Red + white  Pink
Violet + white  Lavender
11
of
36
Color Image Processing
 The amounts of red, green and blue needed to form
a particular color are called tri-stimulus X, Y, Z
 A color is specified by its tri-chromatic coefficients:

 x+y+z=1
12
of
36
Color Image Processing
 Another approach for
specifying colors is to use
CIE (International Commission on
Illumination) chromaticity
diagram which shows color
composition as a
function of x (red) and y
(green).
 Various spectrum colors
are indicated around the
boundary of tongue-
shaped diagram (pure
colors).
13
of
36
Color Image Processing
 Any point not actually on the boundary represents some
mixture of spectrum colors
 Any point on the boundary of chart is fully saturated.
 A straight-line segment joining any two points in the diagram
defines all different colors that can be obtained by combining
these two colors additively.
 A line drawn from the point of equal energy (white) to any
point on the boundary will define all the shades on that color
 To determine colors that can be obtained from any three
given colors, we draw connecting lines to each of the three
color points
 Any color inside the triangle can be produced by various
combinations of the three initial colors
14
of
36
Color Image Processing
 Color Models (Color Space or Color System)
 Color models: used to specify colors in a standard
way
 Color model is a coordinate system where each
color is represented by a single point
 RGB (red, green, blue)
 CMY (cyan, magenta, yellow)
 HIS (hue, saturation, intensity)
15
of
36
Color Image Processing
 Each model is oriented toward a hardware or
application.
 RGB: color monitors, cameras, color image
processing
 CMY: color monitor
 HIS: color image processing
 HIS has the advantage that it decouples color and gray
level information in an image making it suitable for many
gray scale techniques discussed before
16
of
36
RGB
 Each color appears in its primary spectral components of
red, green and blue
 Different colors in this model are points on or inside the
cube
 Number of bits used to represent each pixel in RGB space is
called the pixel depth
 If 8 bit is used to represent each color, 24-bit RGB color
image is obtained
 Total number of colors: (28)3
 A 24-bit RGB
color cube
17
of
36
RGB
 A subset of colors that are likely to be reproduced
reasonably independent of viewer hardware is
called safe RGB color
 216 colors have become the de facto standard for
safe colors
 Each of 216 safe color is formed from three RGB
values but each value can only be 0, 51, 102, 153,
204 or 255.
18
of
36
RGB

(a) Generating the

RGB image of the
cross-sectional color
plane (127, G, B).
(b) The three hidden
surface planes in the
color cube
19
of
36
CMY
 Most devices that deposit colored pigments on paper (color
printers and copiers) require CMY data input
 Equal amounts of pigments primaries, cyan, magenta and
yellow should produce black
 In practice combing these colors produces a muddy looking
black
 In order to produce true black a fourth color black is added
(CMYK) “CMYK = CMY + Black”
When publishers are talking about four color printing, they
are referring to CMY plus black
20
of
36
CMY
K = min (C, M, Y)
 If K = 1, then we have pure black, with no color
contributions, from which it follows that C = M = Y = 0
Otherwise

 Where all values are assumed to be in the range [0, 1]. The
conversions from CMYB to CMY are:
21
of
36
HSI
 RGB and CMY color models are not well suited for
describing colors in terms that are practical for
human interpretation
 HSI (hue, saturation and intensity) decouples
intensity components from the color-carrying
information
 Hue: dominant color
 Saturation: relative purity or the amount of white
mixed with a hue
22
of
36
HSI
 Conceptual relationships between the RGB and HSI color
models.
23
of
36
HSI
 Hue and saturation in the HSI color model. The dot is any color point.
The angle from the red axis gives the hue. The length of the vector is the
saturation. The intensity of all colors in any of these planes is given by
the position of the plane on the vertical intensity axis.
24
of
36
Converting Colors from RGB to HSI
25
of
36
Converting Colors from RGB to HSI
The HSI color model based on (a) triangular, and (b) circular
color planes. The triangles and circles are perpendicular to
the vertical intensity axis.
26
of
36
Converting Colors from RGB to HSI
(a) RGB image and
the components of
its corresponding
HSI image:
(b) hue,
(c) saturation, and
(d) intensity.
27
of
36
Converting Colors from HSI to RGB
RG sector (0≤ H < 120 : When H is in this sector, the RGB
components are given by the equations

GB sector (120≤ H < 240 : When H is in this sector, first

subtract 120from it:
28
of
36
Converting Colors from HSI to RGB
(a)-(c) Modified HSI
component images.
(d) Resulting RGB
image.