1. The document discusses various topics related to digital image representation and processing. It describes how images are digitized through sampling and quantization. 2. Pixel values, image resolution, file formats for storing images like JPEG and GIF are explained. Techniques for image editing like selection tools, painting tools, layers and blending are also covered. 3. The document provides an overview of important concepts in digital image representation and processing including how images are digitized and stored as digital data, techniques for editing images, and methods for manipulating pixels within an image.

1. Lecture # 17
Image Representation
Rahmatullah Danish 1

2. "Be wary, therefore, when
some demand public
tolerance for whatever
their private indulgences
are!"
Rahmatullah Danish 2

3. "Be wary, therefore, when
some demand public
tolerance for whatever
their private indulgences
are!"
- Neal A. Maxwell
Rahmatullah Danish 3

4. Image Representation

5. Digital Image Sources
 Digital Cameras
 Scanned Film & Photographs
 Digitized TV Signals
 Computer Graphics
 Radar & Sonar
 Medical Imaging Devices (X-Ray, CT)
 The Internet
Rahmatullah Danish 5

6. Images - 2D array of values = pixels
 Pixel = “Picture Element”
 Image [x,y] = pixel value (number)
Rahmatullah Danish 6

7. Pixels and Pixel Values
 Pixel – an element of the 2-D image array
 Pixel Value = brightness
 - black = 0
 - gray = 128
 - white = 255
 - many shades over the 0-255 range
0 1 2 3 4 5 6 7
0
1
2
3
4
5
6
7
Rahmatullah Danish 7

8. Digitizing Images
Images are digitized using a two step
process:
1. sampling the continuous tone image
2. quantizing pixels
Rahmatullah Danish 8

9. Sampling
Rahmatullah Danish 9

10. Quantization
pixel’s samples are averaged
Rahmatullah Danish 10

11. Quantization Example
Rahmatullah Danish 11

12. Image Resolution
68 x 104 136 x 208 272 x 416
less detail more detail
less storage more storage
Rahmatullah Danish 12

13. Digital Cameras
 Very Low Res 640x480 (TV grade)
 Medium-Low Res 1024x768
 Medium Res 2048x1536
 Medium-Hi Res 3072x2048
 Hi-Res 3264x2448
Rahmatullah Danish 13

14. Dynamic Range
256 levels
• The number of quantized pixel values:
16 levels 4 levels 2 levelsRahmatullah Danish 14

15. Images - 2D array of values
 Binary Images (pixel values = 0,1)
 Grayscale Images (pixel values = 0-255)
 Color Images
 Each pixel has three color components
 For example, (red, green, blue) or RGB
 Each color component is 0-255
Rahmatullah Danish 15

16. Color Images
3 Images Overlayed
Red
Green
Blue
Rahmatullah Danish 16

17. Histograms: What’s in the image?
 What is a histogram?
 Simple numeric example
1 3 2
13
1
2
1
4
2 1 2 1 2 3 4
HistogramImageRahmatullah Danish 17

18. Color Image Histograms
 Histogram for each color
Rahmatullah Danish 18

19. RGB Additive Color Model
RED
bright values => high amounts of that color
dark values => low amounts of that color
GREEN BLUE
Rahmatullah Danish 19

20. CMYK Subtractive Color Model
Bright => use less of
that ink color
Dark => use lots of
that ink color BLACK
CYAN MAGENTA
YELLOWRahmatullah Danish 20

21. HSB Visual Color Model
HUE
BRIGHTNESS
SATURATION
HSB: how artists perceive color properties
Rahmatullah Danish 21

22. HSB Visual Color Model
HUE
BRIGHTNESS
SATURATION
HSB: how artists perceive color properties
Select Hue
0
o
Hue
360
o
Rahmatullah Danish 22

23. HSB Visual Color Model
HUE
BRIGHTNESS
SATURATION
HSB: how artists perceive color properties
Saturation
Select Hue
0
o
Hue
360
o
- then click in box for saturation, brightnessRahmatullah Danish 23

24. HSB Visual Color Model
HUE
BRIGHTNESS
SATURATION
HSB: how artists perceive color properties
Brightness
Saturation
Select Hue
0
o
Hue
360
o
- then click in box for saturation, brightnessRahmatullah Danish 24

25. Storing Digital Images
 Digital images are converted to files
for storage and transfer
 The file type is a special format for
ordering and storing the bytes that
make up the image
 File types or formats are not
necessarily compatible
 You must often match the file type
with the application
Rahmatullah Danish 25

26. Storing Digital Images
 GIF (Graphic Interchange Format)
 indexed color (up to 256 colors)
 compressed
 used in Web applications
 JPEG (Joint Photographic Experts Group)
 lossy compression with variable controls
 also used in Web applications
Rahmatullah Danish 26

27. Storing Digital Images
 PNG (Portable Network Graphics)
 designed for online viewing (e.g., Web)
 patent-free replacement for GIF
 lossless compression
 BMP
 MS Windows image format
Rahmatullah Danish 27

28. How Many Bytes to Store an Image?
 Suppose we a have an image that is 500x500 pixels in size
 That’s a total of 250,000 pixels
 Binary image (1 bit/pixel) = 31,250 bytes
 Grayscale image (8 bits/pixel) = 250,000 bytes
 Color image (24 bits/pixel) = 750,000 bytes
Rahmatullah Danish 28

29. Indexed Color
 “Indexed Color” can be used to reduce the size of a color image file
0 1 0 1 2 1 0 1 0
0 1 2
255 255 255 255 0 255 255 255 255
0 255 0 255 255 255 0 255 0
0 0 0 0 255 0 0 0 0
= 27 bytes
= 18 bytes
Rahmatullah Danish 29

30. Indexed Color Images
 are derived from full
color images
 are smaller or more
compact in storage
 are composed of
pixels selected from
a limited palette of
colors or shades
Demo: GIMP PosterizeRahmatullah Danish 30

31. Image Processing II
Rahmatullah Danish 31

32. 2 Classes of Digital Filters
 Global filters transform each pixel uniformly according to the function
regardless of its location in the image
 Local filters transform a pixel depending upon its relation to
surrounding ones
Rahmatullah Danish 32

33. Global Filters: REVIEW
 Brightness and Contrast control
 Histogram thresholding
 Histogram stretching or equalization
 Color corrections
 Inversions
Rahmatullah Danish 33

34. Image
Editing
Rahmatullah Danish 34

35. Image Editing
 Selection Tools
 Painting Tools
 Cut & Paste
 Cloning
 Layers and Blending
Rahmatullah Danish 35

36. Selection Tools
 Rectangular Selection
 Oval Selection
 Lasso Tool
 Magic Wand
 Color Select Tool
 Intelligent Scissors
 Foreground Select Tool
DEMOS
Tool Bar
Rahmatullah Danish 36

37. Image Editing
 Selection Tools
 Painting Tools
 Cut & Paste
 Cloning
 Layers and Blending
Rahmatullah Danish 37

38. Painting Tools
 Paint Bucket Tool
 Gradient Shade
 Pencil Tool
 Paintbrush Tool
 Eraser
 Airbrush Tool
 Ink Tool
DEMOS
Rahmatullah Danish 38

39. Image Editing
 Selection Tools
 Painting Tools
 Cut & Paste
 Cloning
 Layers and Blending
Rahmatullah Danish 39

40. Cut & Paste
 Word Processors
- cut & paste strings of characters (1D arrays)
 Image Editing
- cut & paste pixels (2D arrays)
- replace old pixels with new pixels
A s t r i n g
65 32 115 116 114 105 110 103
0 1 2 3 4 5 6 7
Rahmatullah Danish 40

41. Image Editing
 Selection Tools
 Painting Tools
 Cut & Paste
 Cloning
 Layers and Blending
Rahmatullah Danish 41

42. Cloning
 Copy pixels from
one part of an image
- to another part of an image ... Interactively
DEMORahmatullah Danish 42

43. Image Editing
 Selection Tools
 Painting Tools
 Cut & Paste
 Cloning
 Layers and Blending
Rahmatullah Danish 43

44. Layers and Blending
Can create arbitrary number of layers for
- animation
- special effects in movies
- morphing
Layer 1
Layer 2
Layer n
Rahmatullah Danish 44

45. Blending
 The idea: Blended image
= .3 x + .7 x
is a weighted combination (sum) of
two or more other images.
Rahmatullah Danish 45

46. Example Blend
.3 x +.7 x
= Bearastronaut
Rahmatullah Danish 46

47. Masking
 The idea: Create another image
where the value of pixels is the
weighting term for a blend operation:
Rahmatullah Danish 47