ETI 2507 DIGITAL IMAGE

PROCESSING
Colour Image Processing

Khamis Bamama

1/ 22
 Lecture Outline
Lecture Outline
Introduction

LECTURE OUTLINE

1 A recap on previous lecture

Human Visual System
2 Colour image processing
Introduction to Color Image Processing
Human Perception of Color
Color Models
Color Image Representation
Applications of Color Image Processing
Example Application: Fruit Defect Detection

2/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

Introduction to Color Image Processing

Color image processing involves manipulation and analysis of

color images.
Adds significant information for object recognition, scene
understanding, etc.
Typical representation using RGB color channels.

3/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

Human Perception of Color

Human eye has three types of cones: red, green, and blue.
Additive and Subtractive color models.
Additive: RGB, used in screens and displays.
Subtractive: CMY/CMYK, used in printing.

4/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

Human Visual perception

5/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

Colors of Lights & Pigments

Primary Colors of Light

(Additive): R, G, B
R + G → Yellow
R + B → Magenta
G + B → Cyan
R + G + B → White
Primary Colors of
Pigment (Subtractive):
M, C, Y
M ∩ C → Blue
M ∩ Y → Red
Y ∩ C → Green
Y ∩ M ∩ C → Black

6/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

Chromaticity in Color Image Processing

Chromaticity refers to the color quality defined by hue and

saturation, independent of brightness or luminance.
It represents the color’s hue (e.g., red, green, blue) and its
saturation (intensity or purity of the color).
A way of describing the color’s chromatic properties without
considering how light or dark the color is.
Chromaticity is typically visualized in a two-dimensional color
space where luminance is excluded.

7/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

Chromaticity

Let X , Y , Z be Tri-stimulus values representing the amounts

of red, green, and blue needed to form any particular color.
Defined as:
X Y Z
x= , y= , z=
X +Y +Z X +Y +Z X +Y +Z

Normalization: Since x + y + z = 1, x and y alone determine

a chromaticity diagram.
Any color can be specified by x and y alone, known as
chromaticity coordinates.

8/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

Chromaticity Diagram

A 2D representation of color based on hue and saturation,

excluding luminance.
Derived from the CIE 1931 Color Space and maps human
color perception.
The x - and y -coordinates represent color components, with
the white point at the center.
Applications:
Used in color science, color correction, and gamut mapping.
Helps visualize color relationships and color mixing.

9/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

CIE Chromaticity Diagram

10/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

Color Gamut
The color gamut is the
range of colors that a device
or medium can represent or
display.
It is often depicted in the
chromaticity diagram, where
the boundary defines the
gamut.
A larger gamut means a
wider range of colors can be
represented.
Any 3 points in the
chromaticity diagram can
produce all colors within
that triangle.
11/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

Color Models

A color model is a method for specifying colors in a standard

way.
Consists of a three-dimensional coordinate system.
Each color is represented by a single point within a subspace of
the system.
Examples include:
RGB Color Model
CMY (CMYK) Color Model
YIQ/YUV/YCbCr Color Models
HSI Color Model

12/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

RGB Color Model

RGB is the most commonly

used color model.
Additive color model with
Red, Green, and Blue
channels.
Each channel has intensity
values (0-255 for 8-bit
images).
Applications: Monitors,
video

13/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

CMY/CMYK Color Model

Subtractive color model used in printing.

CMY: Cyan, Magenta, Yellow.
CMYK adds a black component for depth in dark colors.
Applications: Printing.

14/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

YIQ/YUV/YCbCr Color Models:

These models separate a color’s luminance (Y ) from its

chrominance (I/Q, U/V , Cb/Cr ) which represent the color
information.
Used in video and image compression to reduce bandwidth
and storage requirements.
Applications:
YIQ: Color TV (NTSC).
YUV: Color TV (PAL).
YCbCr: JPEG, MPEG.

15/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

HSI Color Model

Also known as: HSL, HSV

I: Intensity, L: Lightness, V : Value.

Components: Hue, Saturation, Intensity (Value).

Hue: An attribute describing pure color.
Saturation: The degree to which a pure color is diluted by
white light.
The HSI model is designed to represent colors similar to
human perception.
Useful for color-based segmentation and analysis.

16/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

Color Space Conversion

Converts between color models (e.g., RGB to HSV).

Necessary for tasks like image compression, segmentation.

17/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

Color Coordinate Transform

RGB to CYM:      
C 1 R
 Y  = 1 − G 
     
M 1 B

HSI to RGB (for 0 ≤ H ≤ 120◦ ):

B = I(1 − S)
S · cos H
 
R =I 1+
cos(60◦ − H)
G = 1 − (R + B)

RGB to YIQ:
     
Y 0.299 0.587 0.114 R
 I  = 0.596 −0.275 −0.321 · G 
     
Q 0.212 −0.523 0.311 B
18/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

Practical Applications of Color Image

Processing

Image Enhancement.
Smoothing and Sharpening.
Color Image Segmentation

19/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

Example: Fruit Defect Detection

20/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

Example: Fruit Defect Detection

21/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING
 Lecture Outline
Introduction
Introduction

Next Lecture

Filtering in the spatial domain

22/ 22
Khamis Bamama ETI 2507 DIGITAL IMAGE PROCESSING