Machine Vision

1
VISION

the most powerful sense

2
 Machine Vision System

• creates a model of the real world

from images

• recovers useful information about a

scene from its two dimensional
projections

3
 A typical control system:

Imaging MACHINE
device VISION

Scene Image Description

Illumination
Application
feedback

4
 Components of a M.V. system

HARDWARE
SOFTWARE
• Optics (lenses, lighting)
• Cameras
• Interface (frame grabber)
• Computer

5
 Components of a M.V. system

HARDWARE • Libraries with custom code

SOFTWARE developed using Visual
C/C++, Visual Basic, or
Java
• Graphical programming
environments

-There is no universal vision system

-One system for each application
6
Application Fields
• Medical imaging
• Industrial automation
• Robotics
• Radar Imaging
• Forensics
• Remote Sensing
• Cartography
• Character recognition

7
 Machine Vision Stages
Image Acquisition Analog to digital conversion

Remove noise,
Image Processing improve contrast…

Find regions (objects)

Image Segmentation in the image

Take measurements of
Image Analysis objects/relationships

Match the description with

Pattern Recognition similar description of known
objects (models)

8
 Image Formation
Projection
3D 2D

• Perspective projection

• Orthographic projection

9
 Digital Image Representation
Image: 2D array of gray level or color values

Pixel: Array élément Picture Element

•It`s 2D
•It`s a square
•It has one color
•It may be any color

Pixel value: Arithmetic value of gray level or color intensity

Gray level image: f = f(x,y)

Color image: f = {Rred(x,y), Ggreen(x,y), Bblue(x,y)}
10
 Digital Image Formation

Sampling Quantization

• Sampling rate • Quantization level

(resolution)

11
Different Sampling Rates

12
Different Quantization Levels

13
Image Processing (IP)
Image
Processing

Input Image Output Image

Filtering
Smoothing
Thinning
Expending
Shrinking
Compressing
…
14
 Fundementals
• Neigborhood

4-Neighbors 8-Neighbors

• Histogram: gray levels vs number of pixels

15
 IP Examples

Removal of noise with median filter

16
 IP Examples (2)

Improvement of contrast by histogram modification

17
IP Examples
(3)

Symethrical
inversion of the image

18
 IP Examples (4)

Original Uniform Gaussian

Smoothing

19
 Binary Image Processing
WHY?
better efficiency in acquiring, storage,
processing and transmission

TRESHOLDING

20
Different tresholds

21
 Image Segmentation
Segmentation

Regions
Input Image
Objects

-Clasify pixels into groups having similar characteristics

-Two techniques: Region segmentation

Edge detection

22
 Region Segmentation
• Histogram Based

23
 Region Segmentation (2)

• Split & Merge

24
 Edge Detection
• Find the curves on
the image where
rapid changes occur

25
 Image Analysis
Image
Analysis

Input Image
Measurements
Regions, objects

Measurements:
-Size
-Position
-Orientation
-Spatial relationship
-Gray scale or color intensity
-Velocity
26
 Pattern Recognition (PR)
Pattern
Recognition
- Measurements
- Stuctural Class identifier
descriptions

feature vector
set of information data

27
Approaches to PR

Statistical

Structural

Neural

28
• 3D VISION

• Dynamic Vision

29
Fingerprint Recognition

- most precise identification

biometric
- has many applications
- has the largest database
30
31
 Fingerprint recognition system

Enrollment

Fingerprint Template
Feature Extractor
sensor database

Identification

Fingerprint
Feature Extractor
sensor

Feature Matcher

ID
32
 Fingerprint Representation
Local Ridge Characteristics

•Ridge Ending
•Ridge Bifurcation
•Enclosure
•Ridge Dot

33
Fingerprint Representation (2)
Singular Points
Core
Delta

Major Central Pattern

Arch Loop Whorl 34

 Image Processing & Analysis for
Fingerprint Recognition

Minutiae
Pre-processing Post-processing
Extraction
•Binarization •Ridge break
removal
•Noise Removal
•Bridge removal
•Smoothing
•Spike removal
•Thinning

Processed
Image
+
Minutiae
Description
Input Image 35
Outputs
 Pre-Processing
• Binarization
search through array pixel by pixel;
if current byte colour value is above threshold then
change value to white;
else
change colour to black
• Noise Removal
if the pixel is white and all immediate surrounding pixels are
black
then
change pixel to black;
else if the pixel is black and all immediate surrounding pixels
are white then
change pixel to white;
36
 Pre-Processing(2)
• Smoothing
for each pixel do
add the values of surrounding pixels;
divide by the number of surrounding pixels (usually 8 unless
at the edge);

assign current pixel the calculated result;

• Thinning

37
 Extraction
if pixel has exactly one neighbour then
pixel is a ridge ending;
calculate x value of pixel;
calculate y value of pixel;
calculate angle of ridge ending;
write data to database;

else if pixel has exactly three neighbours then

pixel is a bifurcation;
calculate x value of pixel;
calculate y value of pixel;
calculate angle of bifurcation;
write data to database;

38
 Post-Processing
• Ridge break removal
If A and B are facing each other and are less then a minimum
distance (D) then
A and B are false minutia points;
remove A and B from ridge end point set;
update database;
Bridge removal

• Spike removal
For each ridge ending point (A)
For each bifurcation point (B)
If A and B are less than minimum distance (D)apart
and they are connected then
A and B are false minutia points;
remove A and B from their sets; 39
Matching

40
 Applications
Machine Vision Systems used in Manufacturing and
Robot automation to perform various tasks:
• Inspection
• Part Identification
• Part Orientation
• Part Location
• Visual guidance & Control
• Safety Monitoring

41
 Applications (2)
• Inspection
For quality control of parts.
Includes:
Checking of gross surface defects
Discovery of flaws or mistakes during final
inspection in labeling
Verification of presence of components in
assembly
Measuring for dimensional accuracy
Checking for presence of holes and other features
in parts.
42
 Applications (3)
• Part Identification
Used where vision system stores data for different
parts in active memory.
Used to recognize or distinguish between parts.
Includes:
Part sorting
Part counting
Part character recognition.

43
 Applications (4)
• Part Orientation
Used in parts to be gripped in a specific manner by
the End-effector.
Vision system supplies both information and data to
drive the gripper into correct orientation for pick and
place action.

44
 Applications (5)
• Part Location
Used to locate randomly placed parts on x, y axes.
Vision system measures the x & y distances and
identifies the center of the camera to coincide with
the center of the randomly placed part.

45
 Applications (6)
• Visual Guidance & Control
Vision system is teamed with a robot or other
devices in the robot cell to control the movement of
the machines.
For example:
Seam tracking in continuous arc welding
Part positioning and reorientation
Bin picking

46
 Applications (7)
• Safety Monitoring
Vision system id used to monitor the operation of
production cell.
Purpose is to detect irregularities that indicate a
condition that is hazardous to equipment or people
working in the cell.
Also used to detect intruders entering the place who
might be at risk by wandering into the robot cell.

47
THANK YOU!

48