INTRODUCTION

TO
IMAGE PROCESSING

1
COMPUTER VISION
• Computer vision is a field that includes methods for
• acquiring
• processing
• analyzing
• understanding images
• In general, high-dimensional data from the real world in order to
produce numerical or symbolic information, e.g., in the forms of
decisions.
• A theme in the development of this field has been to duplicate the
abilities of human vision by electronically perceiving and
understanding an image

2
RELATED FIELDS

3
WHAT IS MACHINE VISION?
• Machine vision is the technology and methods used to
provide imaging-based automatic inspection and
analysis for such applications as automatic inspection,
process control, and robot guidance in industry.

• automate the production,

• increase production speed, yield and
• improve the quality of the products.

4
WHAT IS DIGITAL IMAGE PROCESSING?
• A image may be defined as two-dimensional function, f(x, y).
• Where x and y are spatial coordinates.
• The value at any pair of coordinates of f(x, y) is called as intensity
or gray level.
• These elements are called PIXEL elements.
• Enhancing an image
• Extracting information features from an image.
• Pattern recognition

5
6
7
TYPES OF IMAGES
• Binary image

• Gray scale image

• Color image

8
BINARY IMAGE

9
GRAY IMAGE

10
COLOR IMAGE

11
ROLE OF IMAGE PROCESSING
• Improvement of pictorial information for Human interpretation
• Enhance the contrast levels for easier interpretation of X-ray
images
• Removing the noise from aerial and satellite imagery
• Processing scene of data for autonomous machine perception
• Character recognition(OCR,OCV)
• Machine Vision system for assembly and inspection
• Fingerprint processing

“Digital Image Processing plays a vital role in Machine vision

systems”

12
COORDINATE CONVENTION

In many image processing books In the Image ProcessingToolbox

13
STAGES IN DIGITAL IMAGE PROCESSING
Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Representation
Problem Domain
& Description
Colour Image Image
Processing Compression

14
IMAGE ACQUISITION
Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Representation
Problem Domain Colour Image Image & Description
Processing Compression

15
IMAGE ENHANCEMENT
Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Representation
Problem Domain
& Description
Colour Image Image
Processing Compression

16
IMAGE RESTORATION
Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Representation
Problem Domain
& Description
Colour Image Image
Processing Compression

17
MORPHOLOGICAL PROCESSING
Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Representation
Problem Domain
& Description
Colour Image Image
Processing Compression

18
SEGMENTATION Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Representation
Problem Domain
& Description
Colour Image Image
Processing Compression

19
OBJECT RECOGNITION:
Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Representation
Problem Domain
& Description
Colour Image Image
Processing Compression

20
REPRESENTATION & DESCRIPTION
Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Representation
Problem Domain
& Description
Colour Image Image
Processing Compression

21
IMAGE ACQUISITION….
Lighting
device

object
Camera

“Machine Vision system starts with image acquisition followed by

Image Processing”

22
FACTORS AFFECTING LIGHTING
• Lighting device

• Test object

• Lighting environment

• Machine vision software

23
ARRANGEMENT OF LIGHTING
• Incident Light
• Needs incident light reflections or at least scattering of the test object for
imaging.
• Transmitted light
• Positioned at the opposite side of the object plane.
• Creates sharp and good contrast for low opaque or low transparent parts.

24
FACTORS FOR BUILDING MACHINE VISION
SYSTEMS
• Camera scan type
• FOV
• Resolution
• Lens
• Illumination
• Electrical and Mechanical interfaces
• Designing and choosing software

25
CAMERA PARAMETERS

26
CONSIDERATION FOR IMAGE ACQUISITION
• Choosing CAMERA
• Field of view
• Resolution
• Lens
• Lighting source
• Designing and choosing Software
• MATLAB
• Halcon
• openCV

27
 SIGNAL FLOW OF IMAGE FROM CAMERA TO COMPUTER (ANALOG)

Vision Camera
Frame Grabber
Analog Signal
RS 170 / CCIR

DAC Analog Signal ADC

Digital Image Sensor

Image Buffer

28
 SIGNAL FLOW OF IMAGE FROM CAMERA TO COMPUTER (DIGITAL)

Vision Camera
Frame Grabber

Digital Signal
MAY BE, or
becomes a
part of the
camera

DAC Analog Signal ADC

Digital Serial or Parallel

Interface
Digital Image Sensor Image Buffer
29
WHY LIGHTING?

• Light carries the primary information

• Every information comes from the light.
• The better the lighting solution, the better machine
vision system will perform
• Contrast, brightness and darkness , shadows and textures of an image
are done by lighting.

“Perfect lighting makes image processing easier”

30
LIGHT SOURCES
• Incandescent Lamps
• Halogen Lamps
• Metal vapour lamps Halogen
Incandescent
• Xenon lamps
• Fluorescent lamp
• Light emitting diode LED Fluorescent

• Lasers
LED

Xenon
31
LIGHTING COMPONENT

32
33
IMAGE ENHANCEMENT
Image Morphological
Restoration Processing

Image
Segmentation
Enhancement

Image Object
Acquisition Recognition

Image Representation
Problem Domain Colour Image
Compression & Description
Processing

34
IMAGE ENHANCEMENT…
• It is performed to remove unwanted clutter or noise
from an image.
• Process of adjusting digital images so that the results
are more suitable for display or further analysis.
• Technique used to improve the quality of an image.
• Enhance Information

• Make Decision analysis easier

35
 ENHANCEMENT TECHNIQUES

Histogram
Gray level stretching
Processing

Image averaging
Filtering

Image Subtraction

36
HISTOGRAM PROCESSING
• An image histogram is a chart that shows the
distribution of intensities in an indexed or grayscale
image.
• You can use the information in a histogram to choose
an appropriate enhancement operation.
• For example, if an image histogram shows that the
range of intensity values is small, you can use an
intensity adjustment function to spread the values
across a wider range.

37
HISTOGRAM FOR DIFFERENT CONTRAST
• It is the probability of occurrence associated with the gray levels
in the range 0 to 255.

LOW CONTRAST MEDIUM CONTRAST HIGH CONTRAST

38
GRAY SCALE STRETCHING
• Gray scale stretching is achieved by first shifting all the values so that
the actual pixel grey-level range begins at 0.
• Add to every pixel the difference between the final low value(0) and
the initial low value(100): 0-100= -100

39
CONTD….
• Next , we scale everything, reassigning values in the range 0 – 100 to
the range 0 – 255 i.e we scale by a factor = (255-0)/(100-0) = 2.55
• Thus, in this instance, to stretch the contrast so that all the grey-scale
are utilized, one must simply apply the following operation.

40
MORPHOLOGY
• Used for shape and structural manipulations. Can be done in binary
or grayscale images.
• Neighborhood operator that determines each pixel value according to
its geometrical relationship to its neighbors.
• Morphological operator can:
• Thin
• Thicken
• Fill

41
STRUCTURING ELEMENT
• The operations of binary morphology input a binary image B and
structuring element S.
• S is another binary image or sub image.
• It represents a shape, can be of any size and have arbitrary structure that
can be represented by a binary image and depend on application.
• Members and origin of SE are specified.
• A symmetrical one will enlarge or shrink the original set in all directions.
• A vertical one, will only expand or shrink the original set in the vertical
direction.

42
STRUCTURING ELEMENT

43
BASIC MORPHOLOGICAL OPERATIONS
• The basic morphological operations of binary morphology are
dilation, erosion, closing, and opening.
Dilation enlarges a region.
Erosion makes a region smaller.
A closing operation can close up internal holes on a region and
eliminate bays along boundaries.
An opening operation can get rid of small portions of the region that
run out from the boundary.

44
DILATION
• Can be though of as rolling a
1 2
ball along the edge and adding
4 3
pixels where it fits.
• Any pixel that has one or
more of its neighbors white is
turned to white.
Dilate

45
DILATION
+
• Dilation of set B by structuring element S
• New set C=B S is composed of all the points obtained by replacing
every point(x , y) in B with a copy of S , placing the origin point(0,0)of
S at (x , y). The replacement operation works vice versa.
S
Dilating B with a 3X3
structuring element S of
ones, centered at origin

46
DILATION
• Dilate a binary image with a vertical line structuring element.

47
EROSION
• Can be thought of as peeling away layers ,like onion
• Any pixel whose neighborhood is not entirely white is turned black.

1 2
4 3

48
EROSION
• Erosion of set B by structuring element S
• New set C=B S is composed of all the points (x , y) for which S is in B.
This can be done by moving S over B ,find all the places it will fit
completely, and for each such place, mark down the point
corresponding to the origin(0 , 0) point of S.
Eroding B with a 3X3 S
structuring element S of
ones, centered at origin

49
EXAMPLES

50
OPENING
• Opening is erosion followed by dilation

51
CLOSING
• Closing is dilation followed by erosion

52
RGB COLOR SPACE
RGB color image is an MxNx3 array of color pixels , where M, N is
row, column and 3 represents three monochromatic images (R,G,B)
combined to form an RGB color image.
The color image would have 24 bit/pixel – 8 bit for each color bands
(red, green and blue )

53
RGB COLOR SPACE

54
COLOR IMAGES

R M B

W
Y C

Mixtures of Light: Additive Primaries

55
CONTD…
G
B
G
B

R R+G+ R R+G+
B B

B B
G G
R R+G+ R R+G+
B B

B B
G G
R R+G+ R R+G+
B B

56
HSV COLOR SPACE

• The color space other than HSV is not well suited for
describing colors in terms that are practical for
human interpretation.

• Human describe color object by its hue , saturation

and intensity.

57
HSV SPACE
HUE:
• It describes pure color which distinguish red, green, yellow.
• More than 400 hues can be seen by humans.

SATURATION:
• It is measure of a degree to which a pure color is diluted by white light.
• A pure color has 100% saturation and the white ,black have 0% saturation. About 20
saturation levels are visible per hue.

VALUE:
• The average of three layers.

58
CONTD…
Var=rgb2hsv(‘rgb.jpg’);
Hue =Var(:,:,1);
Sat =Var(:,:,2);
Val =Var(:,:,3);

59