Digital Image Processing

Chapter 1 | Introduction
 What is a digital image?

 A digital image is a two-dimensional representation as a finite set of digital values

called pixels.
 Pixel values can represent gray levels, colors, heights, opacities, etc.
 Common image formats include:
o 1 sample per point (B&W or Grayscale)
o 3 samples per point (RGB)
o 4 samples per point (RGBA - with opacity)
 Digitization implies the approximation of a real scene.

 What is Digital Image Processing (DIP)?

 Focuses on two major tasks:

o Improving pictorial information for human interpretation.
o Processing image data for storage, transmission, and representation for
autonomous machine perception.
 It ranges from low-level (noise removal) to mid-level (object recognition) to high-
level processes (scene understanding).

 History of Digital Image Processing

 1920s: Early use in newspaper image transmission (Bartlane system).

 1960s: Space race accelerated digital image processing.
o Used in enhancing moon images from the Ranger 7 probe.
 1970s: Medical applications like CAT scans began.
 1980s - Present: Extensive use across diverse fields like medical imaging, industrial
inspection, law enforcement, and more.

 Examples of Digital Image Processing

 Image Enhancement: Quality improvement and noise removal.

 Hubble Telescope: Image processing corrected faulty mirror issues.

1| P a g e Year 4
Semester 8
  Artistic Effects: To create visually appealing or composite images.
 Medical Imaging: MRI scans for tissue differentiation.
 Geographic Information Systems (GIS): Satellite imagery analysis for meteorology
and terrain classification.
 Industrial Inspection: PCB inspection to detect missing components.
 Law Enforcement: Number plate and fingerprint recognition.
 Human-Computer Interaction (HCI): face and gesture recognition.

 Key Stages in Digital Image Processing

 Image Acquisition: Capturing digital images.

 Image Restoration: Correcting distortions or degradations.
 Morphological Processing: Analyzing structures in images.
 Segmentation: Dividing an image into meaningful regions.
 Representation & Description: Extracting features for further analysis.
 Image Enhancement: Highlighting important features for better visualization.
 Object Recognition: Identifying objects within an image.
 Image Compression: Reducing image size for storage and transmission.
 Color Image Processing: Handling colored images for diverse applications.

Chapter 2 | The Processing Development Environment

 Introduction to Processing Development Environment (PDE)

 Processing is a free, open-source tool with a development environment similar to

Arduino.
 It is primarily based on the Java language, but its code can be translated to other
environments that support serial communication.

 Features of the Processing Development Environment (PDE)

 The PDE has a simple text editor for writing code, a message area, a text console,
tabs for managing files, a toolbar for common actions, and various menus.
 The default mode is Java, but it can change depending on the environment.
 In Processing, a computer program is called a sketch, and sketches are stored in a
folder named Sketchbook on your computer.

2| P a g e Year 4
Semester 8
 Basic Code Examples

 Line Drawing:

void setup() {

size(500, 400);

void draw() {

background(255); // White background

line(60, 50, 100, 50);

 Triangle Drawing:

void draw() {
triangle(60, 10, 25, 60, 75, 65);
line(60, 30, 25, 80);
line(25, 80, 75, 85);
line(75, 85, 60, 30);
}

 Ellipse, Circle and Rectangle Drawing:

void draw() {
ellipse(40, 40, 60, 60); // Circle
rect(50, 50, 30, 40); // Rectangle
}

3| P a g e Year 4
Semester 8
 Quad Literal(Drawing Quadrilateral)

Void draw(){
quad(20, 20, 20, 70, 86, 90, 60, 40);
quad(20, 20, 30, -76, 110, 0, 60, 40);
}

Chapter 3 |Image Processing

1. Loading and Displaying Images in Processing

PImage is used to handle image files in Processing.

PImage img;
img = loadImage("your_image.jpg");
image(img, x, y);

The image must be inside the data folder of your sketch

You can resize images using:

img.resize(width, height);

2. Practical Example 1: Displaying an Image

PImage
PImagehog1;void
hog1;voidsetup(){
setup(){
size(700, 500);
size(700, 500);
hog1 = loadImage("hog.jpg");
hog1
}void = loadImage("hog.jpg");
draw(){
image(hog1,
}void draw(){0, 0);
fill(0, 255, 0); 0, 0);
image(hog1,
ellipse(300, 200, 10, 10);
fill(0, 255, 0);
}
ellipse(300, 200, 10, 10);
}

4| P a g e Year 4
Semester 8
Loads an image and draws a green circle over it.

3. Using Mouse Position for Drawing

You can use mouseX and mouseY for dynamic interactions.

void setup() {
size(200, 200);
}void draw() {
background(255);
rectMode(CENTER);
rect(mouseX, mouseY, 50, 50);
}

The rectangle moves with your mouse.

4. Practical Example 2: Dynamic Image Scaling

PImage images;void setup(){

size(240,240);
images = loadImage("images.jpeg");
}void draw(){
image(images, 0, 0, mouseX,
mouseY);
}

Scales the image based on mouse position (width = mouseX, height = mouseY)

5. Practical Example 3: Tinting Images

tint(r, g, b) adds a color overlay to images.

5| P a g e Year 4
Semester 8
PImage images;void setup(){
size(240,240);
images = loadImage("images.jpeg");
}
void draw(){
tint(255, 0, 0); // Apply red tint
image(images, 0, 0, mouseX,
mouseY);
}

6. Drawing a Character (Zoog)

Uses shapes like rectangles and ellipses to draw a cartoon character.

void draw() {
background(255);
ellipseMode(CENTER);
rectMode(CENTER);
rect(100, 100, 20, 100); // Body
ellipse(100, 70, 60, 60); // Head
ellipse(81, 70, 16, 32); // Left eye
ellipse(119, 70, 16, 32); // Right eye
line(90, 150, 80, 160); // Left leg
line(110, 150, 120, 160); // Right leg
}

7. Drawing Lines from Keyboard Input

 Uses JOptionPane to take user input (coordinates) and draw a line.
 Includes input validation using try-catch.

import javax.swing.JOptionPane;

Draws a line based on entered coordinates.

6| P a g e Year 4
Semester 8
 Chapter 4 |Introduction to MATLAB and image processing
 Introduction

 Image processing involves manipulating images to improve quality or extract

information.
 Digital image processing is widely used in medical imaging, remote sensing, and
industrial automation.
 MATLAB is a powerful tool due to its built-in functions and Image Processing
Toolbox.

 Digital Image Representation

 Images are represented as 2D arrays of intensity values (grayscale) or 3D arrays

(color).
 Each pixel in the array has a value corresponding to brightness or color.
 Color images use three channels: Red, Green, and Blue (RGB).

 Image Types
 Binary Images: Only black and white (0 or 1).
 Grayscale Images: Shades of gray (0–255 for 8-bit images).
 Color Images: Composed of three 2D arrays (RGB).
 MATLAB supports multiple formats: JPEG, PNG, BMP, TIFF, etc.

 Reading and Displaying Images

 imread() is used to read images.
 imshow() displays the image.
 Syntax example:

I = imread('image.jpg');
imshow(I);

 Writing Images

 imwrite() saves images to a file.

7| P a g e Year 4
Semester 8
  Can specify format and filename.
 Syntax example:

imwrite(I, 'output.png');

 Image Conversion

 Convert between types using:

 rgb2gray() – Color to grayscale.
 im2bw() – Image to binary.
 im2double() – Convert to double precision.

 Image Arithmetic Operations

 Operations include addition, subtraction, multiplication, and division.

 Useful for image enhancement or masking.
 Example: Adding two images

C = A + B;

 Image Logical Operations

 Use logical operators: AND, OR, NOT, XOR.
 Useful in segmentation and masking.
 Example:

C = A & B;

 Image Histogram
 A histogram shows pixel intensity distribution.
 imhist() displays the histogram.
 Histogram equalization using histeq() improves contrast.

 Image Filtering

8| P a g e Year 4
Semester 8
  Filtering enhances or suppresses image features.
 Types:

1. Linear Filters (mean, Gaussian)

2. Non-linear Filters (median)
3. Use imfilter() or medfilt2() in MATLAB.

 Edge Detection

 Detects object boundaries.

 Techniques: Sobel, Prewitt, Canny.
 Use edge() function:

BW = edge(I, 'canny');

 Morphological Operations

 Operates on binary images.

 Common operations: dilation, erosion, opening, closing.
 Functions: imdilate(), imerode(), imopen(), imclose().

 Image Segmentation

 Divides image into meaningful regions.

 Techniques: thresholding, edge-based, region-based.
 MATLAB functions: imbinarize(), bwlabel(), regionprops().

Digita Image Processing

Chapter 1, 2, 3, 4
END

9| P a g e Year 4
Semester 8