Intensive Program on Computer Vision

jabgoldfishΤεχνίτη Νοημοσύνη και Ρομποτική

19 Οκτ 2013 (πριν από 4 χρόνια και 2 μήνες)

117 εμφανίσεις

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

1


Intensive Program on Computer Vision

IPCV 200
2



July
22


August 2
, 200
2


Koblenz, Germany


http://www.uni
-
koblenz.de/~lb/lb_activities/ipcv02/ipcv02.html




LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

2

Feature
Extraction

for

Classification
:
Hough Transform and Gabor Filtering


Heikki Kälviäinen

Professor

Computer Science

Laboratory of Information Processing

Heikki.Kalviainen@lut.fi

http/www.it.lut.fi/~kalviai

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

3

Lappeenranta University of
Technolog
y, Finland

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

4

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

5

Contents


Fundamentals of computer vision


Digital image processing


Pattern recognition & Machine vision


Fundamental steps in image processing


Applications


Feature Extraction for Classification


Hough Transform


Gabor Filtering

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

6

Digital Image Processing


R. C. Gonzalez & R.E. Woods, Digital Image
Processing, Addison
-
Wesley, 1993 : “A digital
image is
an image f(x,y)

that has been discretized
both in spatial coordinates and brightness”


f(x,y) is a 2D intensity function where x and y are
spatial coordinates and the value of f at any point
(x,y) is proportional to the brightness of the image
at the point

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

7

Digital Image Processing


A digital image consists of pixels (also called
image elements, picture elements)


For example: an image of a 256 x 256 array with
256 gray
-
levels (8 bits: 0 black, 255 white)


Binary images: only two values


Gray
-
level images: e.g. 256 values


Color images: three color components (e.g. RGB)


Spectral images: several components

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

8

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

9

Pattern Recognition and Machine Vision


A digital image is just a set of pixels ?


Pattern recognition = measurements and
observations from natural scenes and their
automatic analysis and recognition


Computer vision = image analysis using pattern
recognition techniques


Machine vision = application oriented image
analysis

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

10

Fundamental Steps in Image Processing


Image acquisition


Preprocessing


Segmentation


Representation and description


Recognition and interpretation



Image processing system

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

11

Robot Vision: Handling of Sheets in a Workshop

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

12

Robotized Handling of Objects

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

13

Automatic Cheese
Factory
(RTS
, Ltd.
)
Video

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

14

Requirements
for

Successful Applications


Fast



No delays


Accura
te



Assist/replace human vision


Not too expensive



R
eturn on investment


Easy to implement and to use



End users are experts in their own field only!

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

15

Applications (some areas)


Recognition, classification, and tracking of objects


Face recognition, fingerprint detection


Speech recognition, motion detection


OCR, document processing, image databases


Industrial applications



Visual quality control


Process automation


Robotics

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

16

Applications (some areas)



Telecommunications



Image compression, video technology
.



Military applications



Tracking of objects, surveillance systems
.


Remote Sensing


Analysis of satellite images, classification of
airplanes,spying, weather forecasts, forest fire
detection, missile control
.


LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

17

Applications (some areas)



Medical image processing



X
-
ray images, ultrasound images, images of cells,
chromosomes, proteins
.


Detection of tumors, cancer; assistance in operations
.





Chemistry, Biology, Physics, Astronomy



DNA, molecules, particles, planets
.

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

18

Application
s in Finland


TEKES technology programs


Machine Vision (1992
-
1996) & Intelligent and
Adaptive Systems Applications (1995
-
2000) &
Intelligent Automation Systems (2001
-
2004)


Applications of


process control


robot vision


quality control


in
electronics,
metal,
forest
, food manufacturing,
etc., industry & applications in business

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

19

Visual Quality Control in Steel Manufacturing

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

20

Robot Positioning: Deflection Compensation

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

21


Visual Inspection
on Wooden Surfaces

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

22


Visual Inspection
on Wooden Surfaces

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

23

Other Applications


Industrial Robot for
Windscreen Grinding


Quality Control in
Printing Industry


Punch Press Quality
Assurance


Classification of Parquet
Pieces


Controlled Wood Cutting



Automatic Cheese
Production


Detection of Food Fatness


Baking Better Biscuits


Sorting Ceramic Tiles


Multispectral Video


Image databases (see, e.g.
PICSOM,
http://www.cis.hut.fi/resea
rch/demos.shtml)

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

24

References


R. C. Gonzalez & R.E. Woods, Digital Image Processing,
Addison
-
Wesley, 1993.


See more references for example at
http://www.it.lut.fi/opetus/99
-
00/010588000/refs.html


Applications:


Finland: Machine Vision 1992
-
1996. TEKES Technology
Programme Report 15/96. Final Report, 1996.


LUT: http://www.it.lut.fi/research/ip/appl.html


Systems: for example, RTS Group (www.rts
-
group.com)

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

25

Hough Transform


A method for
global

feature extraction:


y = a x + b => b =
-
x a + y.


For each
pixel
(x,y) compute a curve b =
-
x a + b

into the parameter space
.


Alternatively the normal presentation of a line:


Hough Transform detects sets of pixels which represent geometric
primitives in a binary image.


Lines, circles, ellipses, arbitrary shapes, etc.


Tolerant to noise and distortions in an image, but traditional versions
suffer from problems with time and space complexities.


New variants: probabilistic and deterministic Hough Transforms.

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

26

Hough
Transform

(SHT)

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

27

The Kernel of the Hough Transform

1.
Create the set
D

of all edge points in a binary picture.

2.
Transform each point in the set
D

into a parameterized curve in the
parameter space.

3.
Increment the cells in the parameter space determined by the

parametric curve.

4.
Detect local maxima in the accumulator array. Each local maximum
may correspond to a parametric curve in the image space.

5.
Extract the curve segments using the knowledge of the maximum
positions.


LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

28

Randomized
Hough
Transform
(RHT)



Developed in
Lappeenranta University

of Technology (LUT),

FINLAND.



Xu, L., Oja, E.,
Kultanen, P, ”A New
Curve Detection
Method: Randomized
Hough Transform
(RHT),
Pattern
Recognition Letters
, vol.
11, no. 5., 1990, pp.
331
-
338.

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

29

The Kernel of the Randomized Hough
Transform (RHT)

1.
Create the set
D

of all edge points in a binary edge picture.

2.
Select a point pair
(d_i, d_j)

randomly from the set
D
.

3.
If the points do not satisfy the predefined distance limits, go to Step

2; otherwise continue to Step

4.

4.
Solve the parameter space point
(a, b)

using the curve equation with
the points
(d_i, d_j)
.

5.
Accumulate the cell
A(a, b)

in the accumulator space.

6.
If the
A(a, b)

is equal to the threshold
t
, the parameters
a

and
b

describe the parameters of the detected curve; otherwise continue to
Step 2.

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

30



1.
Infinite scope parameter space
.

2.
Arbitrarily high parameter resolution
.

3.
High computational speed
.

4.
Small
storage.


Advances of RHT over SHT

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

31



RHT Extensions


Kälviäinen, H.,

Hirvonen, P.,

Xu, L.,

Oja, E.,

”Pro
b
abilistic and

Non
-
probabilistic

Hough Transforms:

Overview and

Comparisons,”

Image and Vision

Computing
,

Vol. 13, No. 4, 1995,

pp. 239
-
251.

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

32

Feature extraction using Hough
T
ransform


LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

Applications of
Hough Transform


Randomized Hough Transform (RHT)


Curve detection


Motion detection


Mixed pixel classification


Image compression


Vanishing point detection


Image databases


etc.

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

Application of Hough Transform for image
databases


Content
-
based matching of line
-
drawing
images using Hough Transform


Similarity of images in image databases


Hough Transform as a feature extractor


Translation
-
, rotation
-
, and scale
-
invariant
features from the accumulator matrix


LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

35

Compression, Similarity,

Matching, Object Recognition

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

Query images

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

Test database

LAPPEENRANTA

UNIVERSITY OF TECHNOLOGY

THE DEPARTMENT OF INFORMATION TECHNOLOGY

H. Kälviäinen, IPCV 2002, July 22
-

August 2, 2002, Koblenz, Germany

38

Image Processing Using Gabor Filtering


For local and global feature extraction.


Filtering in time (spatial) space and frequency space.


For image processing and analysis two important
parameters: frequency
f
and orientation
theta
.


More information:


Gabor lecture notes
1: (IPCV2002_Gabor1.ps)



Introduction to the theory of Gabor functions.


Gabor lecture notes 2:
(IPCV2002_Gabor2.ps)


Image analysis using Gabor filtering: practice and applications.