Machine Vision lecture 4

1
Machine Visionlecture 4

• Topics
• Feature extraction
• Classification
• Pattern recognition

2
Fundamental Steps in Computer Vision

• Today
• Representation
• Recognition

3
Machine Vision

• Improve image quality
• Binarize image
• Remove noise
• Extract BLOBs
• Calculate features

Today
Feature vector 2,1,,3
. . .
. . .
Feature vector 4,7,,0
4
Machine vision applications

• Pose estimation
• Pick and place applications
• Bin-picking

• Classification
• Quality control

HOF
THOR
Conveyor belt
5
Agenda

• BLOBs
• Finding the pixels for each object
• Extracting features
• Calculate features
• For example size and shape

Feature vector 2,1,,3
. . .
. . .
Feature vector 4,7,,0
6
BLOB what is it?

• Blob versus BLOB
• BLOB Binary Large Object
• Also know as a Particle

. . .
7
Isolating a BLOB

• What we want
• For each object in the image, a list with its
  pixels
• How do we get that?
• Connected component analysis
• (Region growing)
• Define connectivity
• Who are my neighbors?
• 4-connected
• 8-connected

8
Connected component analysis

• Binary image (0,1)
• Seed point where do we start?
• Grassfire concept
• Delete (burn) the pixels we visit
• Visit all CONNECTED (4 or 8) neighbors

9
Connected component analysis
x
x
x
x
x
x
10
Connected component analysis
x
x
x
x
x
x
x
x
x
x
x
x
11
Connected component analysis
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x

• Recursive algorithm
• (show blob, TH, analyze (set outlines))
• Region growing

12
Extracting features
13
Why feature extracting?

• Generate some entities (numbers) that can be used
  for classification
• For example size and shape
• From image operations to mathematical operations
• Input a list of pixel positions
• Output Feature vector
• First step remove too small, too big, and border
  BLOBs
• (show AuPbSn40, TH, analyze (set outlines,
  size, border))

Feature vector 2,1,,3
. . . . .
. . . . .
Feature vector 4,7,,0
14
Features
15
Features
One BLOB

• Area (number of pixels)
• Used to remove noise (small/big objects)
• Number of holes in the object
• Holes area
• Total area area holes area
• Perimeter length of contour
• Bounding box
• Upper left corner
• Height and width of bounding box

16
Features

• Center of mass (xm,ym)
• Compactness
• Minimum for a disc
• Minimum for a rectangle
• Circularity
• Longest distance (Ferets diameter)
• Orientation of the object
• Orientation of Ferets diameter

17
Features

• Shapes
• Bounding box ratio Height / Width
• Says something about the elongation
• How well does the object fit a rectangle
• Can be derived from Area and Perimeter
• How well does the object fit an ellipse
• Can be derived from Area and Perimeter
• Moments
• Many other features exist
• see the notes for today and be creative!

Degree of elongation
18
Classification
19
Classification
THOR
HOF

• Extract some features from each incoming object
• Compare it with learned models
• How?
• Measure the DISTANCE from each BLOB to the model
  and pick the model with the shortest distance,
  i.e. the model most similar to the BLOB
• How do we define a distance?

20
Classification

• Distance in feature-space
• Feature 1 Area
• Feature 2 Circularity
• 2 dimensional feature space
• BLOB
• Models

Feature 2
MATCH!
Feature 1
21
Classification Distance
Feature 2

• Euclidean distance

Feature 1
22
Classification

• Problem
• Area is measured in 1000s and circularity 0,1
• That means that the area will dominate the
  distance measure completely!
• Solutions
• Use ratios So W/H is a better feature than W or
  H
• Normalize all feature to the same interval, e.g.,
  0,1

23
Classification

• What if we trust some features more than others?
• For example, with small objects the perimeter
  might be uncertain
• Solution Weigthing

24
What to remember

• BLOB Binary Large OBject
• Find a list of pixels for each object
• Connectivity 4 versus 8 connected
• Connected component analysis
• Features (characteristics) of an object
• Many exist
• Many more can be derived
• Feature matching
• Make a model of each object to be classified
• Euclidean distance
• Normalize
• Weight

25
Advanced classification
26
Advanced classification

• Distance in feature-space
• Feature 1 Area
• Feature 2 Circularity
• 2 dimensional feature space
• BLOB
• Model types

Feature 2
MATCH!
Feature 1
(tavle note 1-3)
27
What to remember

• Advanced classification
• Based on statistics Bayes classifier
• Include the variance (covariance), since data
  will never be static
• Use enough samples when learning the models!

28
Machine vision applications
29
Machine vision applications

• Pose estimation
• Pick and place applications
• Bin-picking

• Classification
• Quality control

HOF
THOR
Conveyor belt
30
Kvalitets kontrol

• Mål en eller flere features
• Hvordan sætter vi grænserne for accept ?
• Absolutte grænser for samle dele
• Ellers træn grænserne !!
• Find fordelingerne for både OK samt
• FEJL emner (ikke sikkert FEJL er pæn)
• Brug nok samples !

Pointer 1) to typer fejl. 2)
(tavle note 4-5)
31
What to remember

• Quality control
• Learn the thresholds between OK and ERROR!
• Set threshold according to False negatives and
  False positives
• Learn variance of features gt
• Learn covariances gt
• The red region is the objects what will be
  wrongly accepted if ignoring the covariance

32
Xtra
33
Classification

• What if we trust some features more than others?
• For example, with small objects the perimeter
  might be uncertain
• Solution Weigthing
• Advanced classification
• Use more samples to represent a model gt mean and
  covariance
• Pattern recognition
• Gaussian distribution gt
• Mahalanobis distance