National University of Sciences and

Technology (NUST)
School of Electrical Engineering and Computer Science
(SEECS)

Digital Image Processing

Khawar Khurshid 1
 Introduction

Khawar Khurshid 2
 Schedule

• Class Hours

– Mondays 2:00pm - 3:00pm

– Thursdays 11:00am - 1:00pm

• Office Hours

– Thursdays 3:00pm - 4:00pm

or By Email

• Attendance
- Minimum 75% to take the final exam.

Khawar Khurshid 3
 Distribution - Tentative

• Quizzes
10%
• Assignments
5-10%
• OHTs
30%
• Project
10-15%
• Final
40%

Khawar Khurshid 4
 ADIP - Books

Khawar Khurshid 5
 Overview

• Basic Level
– Image Acquisition, Image Representation
– Image Digitization, Quantization
– Morphological Operations, Color Corrections
– Enhancements, De-noising, Sharpening

• Mid Level
– Segmentation, Registration
– Compression, Data Hiding
– Image Restoration, Blur Correction
– Image Transforms

• High Level
– Feature Detection, Feature Analysis
– Intro to Biomedical Imaging
– Classification
Khawar Khurshid 6
 Outline (Subject to Modifications)

• Image Fundamentals
• Sampling, Quantization, Interpolation
• Intensity Corrections, Color Processing
• Morphological Operations
• Transforms
• Restoration
• Compression
• Segmentation
• Image Analysis
• Feature Detection and Analysis
• Biomedical Imaging
• Pattern Classification

Khawar Khurshid 7
 DIP Community

• Conferences
– ACM - Siggraph, Siggraph Asia
– IEEE - International Conference on Image Processing (ICIP)
– IEEE - Computer Vision and Pattern Recognition (CVPR)
– IEEE – International Conference on Pattern Recognition (ICPR)

• Journals
– IEEE Transactions on Image Processing
– IEEE Transactions on Pattern Analysis and Machine Intelligence

– Specific Journals
• IEEE Transactions on Medical Imaging
• IEEE Transactions on Information Forensics and Security

Khawar Khurshid 8
 Key Elements in DIP

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Representation
Acquisition & Description

Object
Problem Domain
Recognition

Colour Image Image

Processing Compression

Khawar Khurshid 9
 Key Elements in DIP

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Representation
Acquisition & Description

Object
Problem Domain
Recognition

Colour Image Image

Processing Compression

Khawar Khurshid 10
 Key Elements in DIP

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Representation
Acquisition & Description

Object
Problem Domain
Recognition

Colour Image Image

Processing Compression

Khawar Khurshid 11
 Key Elements in DIP

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Representation
Acquisition & Description

Object
Problem Domain
Recognition

Colour Image Image

Processing Compression

Khawar Khurshid 12
 Key Elements in DIP

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Representation
Acquisition & Description

Object
Problem Domain
Recognition

Colour Image Image

Processing Compression

Khawar Khurshid 13
 Key Elements in DIP

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Representation
Acquisition & Description

Object
Problem Domain
Recognition

Colour Image Image

Processing Compression

Khawar Khurshid 14
 Key Elements in DIP

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Representation
Acquisition & Description

Object
Problem Domain
Recognition

Colour Image Image

Processing Compression

Khawar Khurshid 15
 Key Elements in DIP

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Representation
Acquisition & Description

Object
Problem Domain
Recognition

Colour Image Image

Processing Compression

Khawar Khurshid 16
 Key Elements in DIP

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Representation
Acquisition & Description

Object
Problem Domain
Recognition

Colour Image Image

Processing Compression

Khawar Khurshid 17
 Key Elements in DIP

Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Representation
Acquisition & Description

Object
Problem Domain
Recognition

Colour Image Image

Processing Compression

Khawar Khurshid 18
 Applications
Of
Image Processing

Khawar Khurshid 19
 Area Classification

Khawar Khurshid 20
 Geographic Information System

 Manipulation of Satellite Imagery

 Terrain Classification

 Weather Imaging

Khawar Khurshid 21
 Digital Elevation Model

Khawar Khurshid 22
 Image Segmentation

Khawar Khurshid 23
 Industrial Automation

Khawar Khurshid 24
 Medical Imaging

Khawar Khurshid 25
 Face Detection

Khawar Khurshid 26
 Face Tracking

Khawar Khurshid 27
 Morphing

Khawar Khurshid 28
 Morphing

Khawar Khurshid 29
 Sign Language/Gesture Recognition

British Sign Language Alphabet

Khawar Khurshid 30
 Lip Reading

Khawar Khurshid 31
 Emotion Classification

 Implicit customer feedback

Khawar Khurshid 32
 Biometrics

 Physiological Biometrics
 Face, IRIS, DNA, Finger Prints

 Behavioral Biometrics
 Typing Rhythm, Handwriting, Gait

Khawar Khurshid 33
 Text Recognition

Khawar Khurshid 34
 Automated Mail Sorting

Khawar Khurshid 35
 Content Based Video Retrieval

Khawar Khurshid 36
 Common Operations on Images

Khawar Khurshid 37
 Common Operations on Images

Khawar Khurshid 38
 Common Operations on Images

Khawar Khurshid 39
 Common Operations on Images

Khawar Khurshid 40
 Common Operations on Images

Khawar Khurshid 41
 Common Operations on Images

Khawar Khurshid 42
 Common Operations on Images

Khawar Khurshid 43
 Introduction

• An image is a 2-dimensional function fxy where x and y

are the spatial coordinates, and fxy is the intensity of
the image at that point.

• A digital image is the representation of a continuous

image f(x,y) by a 2-d array of discrete samples. The
amplitude of each sample is quantized to be
represented by a finite number of bits.

• Each element of the 2D array of samples is called a

pixel (for ‘picture element’)
– Color Images have 3 values per pixel.
– Monochromatic Images have 1 value per pixel.

Khawar Khurshid 44
 Types of Images

– Binary

– Gray Scale

– Color

– Indexed

Khawar Khurshid 45
 Image Storage Requirements

Khawar Khurshid 46
 Intensity Resolution

Khawar Khurshid 47
 Intensity Resolution

Khawar Khurshid 48
 Image Representation

Khawar Khurshid 49
 RGB -> Gray -> Binary
B = im2bw(G, level);

Level = 0.1

I = imread(‘rain.jpg’) I -> 480x720x3

Level = 0.3

Level = 0.5

Khawar Khurshid
G = rgb2gray(I); 50
 R-G-B Color Components

G = I;
G(:,:,1) = 0;
G(:,:,3) = 0;

Khawar Khurshid 51
 R-G-B Color Components

RGB -> RBG

I(:,:,2) = B(:,:,3);
I(:,:,3) = G(:,:,2);
Khawar Khurshid 52
 Resolution

• How many samples and gray levels are required for a

good approximation?

• Quality of an image depends on number of pixels and

number of gray-levels.

• The more these parameters are increased, the closer

the sampled/quantized array approximates the original
image.

• But storage and processing requirements increase

rapidly as a function of N, M, and k.

Khawar Khurshid 53
 Resolution

Depends on what is in the image and what you would like

to do with it.

The picture on the right is fine for counting the number of cars,
but not for reading the number plate.

Khawar Khurshid 54
 End
Introduction

Khawar Khurshid 55