+ All Categories
Home > Documents > HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection...

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection...

Date post: 18-Dec-2015
Category:
Upload: marcus-rose
View: 214 times
Download: 2 times
Share this document with a friend
Popular Tags:
27
HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša Babačev, Marko Barjaktarović University of Belgrade, Faculty of Electrical Engineering Desanka Radunović University of Belgrade, Faculty of Mathematics Belgrade, Serbia and Montenegro Bremen, Germany, 23.01.-26.01.2006.
Transcript
Page 1: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

1

Detection of Cardboard Faults during the Production Process

Nataša Babačev, Marko BarjaktarovićUniversity of Belgrade, Faculty of Electrical Engineering

Desanka RadunovićUniversity of Belgrade, Faculty of Mathematics

Belgrade, Serbia and Montenegro

Bremen, Germany, 23.01.-26.01.2006.

Page 2: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

2

Introduction• Production of cardboard in a long bolt• Occurrence of faults and stains on the surface• Detection and localization in real-time• Existing algorithm using Kirsch operator• Discrete Wavelet Transform using

third level Haar wavelets• Denoising and selection of

vertical wavelet coefficients

Page 3: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

3

Cardboard Production

• Cardboard exits the production machine

• Optoelectronic system photographs the surface

• 800*1024 pixels with 256 levels of gray

• Influence of the factory lights and optic lance

• Nonuniform distribution of gray

• Preprocessing of the image in order to get almost uniform distribution

Page 4: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

4

Original image of cardboard

Page 5: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

5

Almost uniform distribution

Page 6: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

6

Image with fault and noise

Page 7: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

7

Existing algorithm• Wavelets represent the optimization of an

existing algorithm by Marko Barjaktarović• High level of noise

due to short time of exposition andhigh speed of the cardboard

• Denoising is done by a filter of size 1x5 pixels• Extracting of the edges of faults is done by

using the modification of Sobel operator,i.e. Kirsch operator – 5x5 matrix

Page 8: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

8

Result of the Kirsch operator

Page 9: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

9

Converting to binary image

• The threshold is computed from the histogram

Page 10: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

10

Denoising using erosion• Clearing the image of dots that

Kirsch operator detects as an edgebecause of local variations of gray

• The value of every pixel is replaced with the minimum value of neighboring pixels with the same -coordinate and or in order to preserve the line faults that occur

• Two consecutive erosions are needed

x 1x 1x

Page 11: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

11

Erosion of the line fault

Page 12: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

12

Line faults• For a fault in a shape of thin line the result is

more dots with the same -coordinate• Distribution of area of all faults – objects

on the image on -coordinate

• If the area is large for a narrow value of then it is considered a line fault

x

xx

Page 13: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

13

Distribution of area of a line fault

Page 14: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

14

Optimization of the algorithm

• Denoising is done prior the edge detection

• Edge detection

• Both are done using third level Haar wavelets

Page 15: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

15

Denoising with DWT level 3 Haar

• Soft thresholding

• Fixed form threshold t

• s – median absolute deviation of detail coefficients of scale 1

2log(1024*800)t s

Page 16: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

16

Denoised image with level 3 Haar

Page 17: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

17

Extracting the edges

• DWT using third level Haar wavelets

• Faults occur in the direction of motion, i.e. vertical direction

• Only vertical detail coefficients are kept

Page 18: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

18

IDWT with vertical coefficients

Page 19: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

19

Binary values of pixels

Page 20: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

20

Comparing the two methods• Denoised image with DWT is clearer

then in the existing algorithm

• Faster then detecting edges by Kirsch operator

• Less data stored in a sparse matrix

• After IDWT image is with less noise

• Erosion is still needed after IDWT

Page 21: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

21

Image without faults after IDWT

Page 22: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

22

Most common example of fault

Page 23: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

23

Denoising with DWT level 3 Haar

Page 24: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

24

Binary image of IDWT

Page 25: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

25

Image after two erosions of IDWT

Page 26: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

26

Further optimization

• The algorithm using wavelets for edge detection is still in development

• Possible optimization – leaving less vertical coefficients as non-zero

• The right threshold needs to be determined

Page 27: HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory” 1 Detection of Cardboard Faults during the Production Process Nataša.

HASSIP/DFG-SPP1114 Workshop “Recent Progress in Wavelet Analysis and Frame Theory”

27

Detection of Cardboard Faults during the Production Process

Nataša Babačev, Marko BarjaktarovićUniversity of Belgrade, Faculty of Electrical Engineering

Desanka RadunovićUniversity of Belgrade, Faculty of Mathematics

Belgrade, Serbia and Montenegro

Bremen, Germany, 23.01.-26.01.2006.


Recommended