26
Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 ceramics Leszek Chmielewski, Mariusz Nieniewski, Marek Skłodowski, Waldemar Cudny Division of Vision and Measurement Systems (PSWiP) Institute of Fundamental Technological Reserach, PAS (IPPT PAN) Adam Jóźwik Institute of Biocybernetics and Biomedical Engineering, PAS (IBIB PAN)

Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

Embed Size (px)

Citation preview

Page 1: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002

Classification of defectson the surface of black ceramics

Leszek Chmielewski, Mariusz Nieniewski,Marek Skłodowski, Waldemar Cudny

Division of Vision and Measurement Systems (PSWiP)

Institute of Fundamental Technological Reserach, PAS(IPPT PAN)

Adam JóźwikInstitute of Biocybernetics and Biomedical Engineering, PAS

(IBIB PAN)

Page 2: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

2/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Schedule

Objects and their defects Detection of defects Classification of defects Training of the classifier Postprocessing Performance of the processes Results

AcknowledgementsThis research was partly supported by the European Commission: COPERNICUS grant CRASH no. COP - 94 00717 (1995-96) INCO-COPERNICUS grant SQUASH no. ERBIC 15CT 96 0742 (1997-98)

Page 3: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

3/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

This is not concrete – this is ferrite

Black ceramics: ferrite cores magnets

The material is: milled molded pressed sintered ground transported ...

A large number of various defects can emerge during these processesA pair of ferrite cores

Page 4: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

4/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Objects and their defects – called nicely: irregularities

Surfaces:3 important types of defects: crack chip pull-out

Sometimes difficult to classify even for humans

Tiring quality inspection

Page 5: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

5/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Special illumination: controlled light

Tangential, multidirectional light amplifies the visibility of defects

Brightness uniform and independent on distance

LED illuminator Fluorescent illuminator

Page 6: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

6/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Locating the object in the field of view

Simple morphological operations help to find the region occupied by the object for further processing

Aim: to eliminate bright spots and blobs

In this application a narrow stripe at the edge was excluded from analysis

original thresholded complemented

original complemented

Page 7: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

7/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Detection of irregularities (not defects!)

Region of interest is further limited to the irregular part of the surface with the morphological methods

tomorrow’s presentation by prof. Mariusz Nieniewski

original thresholded elongated irregular summed

Page 8: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

8/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Classification of irregularities – features (1/2)

Each pixel detected in the detection phase is classified with the pattern recognition methods. Pixel = pattern.

Features are calculated for each pixel: functions on pixel neighbourhoods = masks.

Direction invariance of features is obtained by rotation of the mask according to local directionality of texture.

Pixel & its mask

original mask rotated mask

tan( )2 2

2

2 2

IxIy

Ix

Iy

dxdy

dxdy

[YBF95]

Page 9: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

9/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Classification of irregularities – features (2/2) brightnesses in the original & rotated mask statistical moments of order up to R in masks gradient modulus 9 classical textural features according to [Law80, Pra91] textural features based on coocurrence relations [WuCh92] relative values of brightness function along the red line

From 30 to 150 features were used for feature selection.

;;

121

242

121

121;;

101

202

101

121;

121

242

121

361

4542321TT AAAAAAA

.

101

000

101

41;

121

242

121

41;;

121

000

121

41

98676

AAAAA T

For example: Features as in [Law80, Pra91]: convolve the mask with A1 – A9 and take standard deviation of the output values 9 features.

Page 10: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

10/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Classification: the K Nearest Neighbour (k-NN) method

Page 11: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

11/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

K-NN: Enhancements and speed-ups

With full selection of features and K Leave-one-out method

Fuzzy version Fuzzy decisions made crisp in the end

Parallel Distinct classifiers for each pair of classes

Hierarchical Advanced version only where classes overlap

Reference set largely reduced with the modified, bidirectional Hart algorithm

Optimized, low error rate, quick algorithm

Page 12: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

12/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

The parallel version of the K Nearest Neighbour method

Page 13: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

13/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

K-NN: class overlap as the training criterion (not error)

Page 14: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

14/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Training: the training patterns

Note: artificial, boundary classes introduced better accuracy

2479 training patterns can be obtained quite quickly...

Page 15: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

15/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Training: first results

The system has successfully classified thousands of unknown pixels.

Quite satisfactory results can be obtained with just 4 training images.

2479 training patterns

raw classifi-cation results

enhanced by local votong

Page 16: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

16/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Training: first results – zoom – results of the enhancement

This was only a convincing example. The error rate estimated with the leave-one-out method was 3.3%.

More training patterns were used in the final system.

pixels used in training classified pixels: all / raw / enhanced

Page 17: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

17/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Final training results – error estimates

5903 training patterns a posteriori error probabilities: pixel classified as class "i"

(row) comes in fact from the class "j" (column); in % overall error: 2.56% max error: 9% between classes 8 and 9

cared for by the postprocessing (to some extent)

Page 18: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

18/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Classification results – various types of elements (1/4)

blue – chip, yellow – good object

Page 19: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

19/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Classification results – various types of elements (2/4)

brown – irregular,

red – crack,

green – pull-out,

blue – chip,

grey – good object.

Page 20: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

20/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Classification results – various types of elements (3/4)

blue – chip,

navy – chip near

good,

red – crack.

Page 21: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

21/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Classification results – various types of elements (4/4)

blue – chip,

green – pull-out,

red – crack !?

Page 22: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

22/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Higher level – discern cracks from grinding grooves

Classify cracks – red betweenall irregular regions – green.

Limits of the method reached.details in other images

Page 23: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

23/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Classification results – rotation (in)variance

Page 24: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

24/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Technical data & Performance

Resolution

A 512*512 pixel camera. Spatial resolutions: 0.05mm/pixel

(up to 20*larger magnifications can be attained with normal lenses)

Accuracy of results

Classification errors: overall up to 4%, inter-class typically 4%, max 10%;

Stability of results

• Detection phase: repeatability not worse than 2-5% in area.

• Classification phase: repeatability not worse than 10-20% in area, depending of how fast classifier version is used.

Processing time [s] (PC, 1000 MHz)

only software morphological processor

image acquisition 0.05-0.20

detection 2.00 0.0001-0.001

classification 1.00 typically; 10 for v. large defects - 20% of object

decision 0.1

Page 25: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

25/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

Conclusion

Irregularities of flat surface in black ceramics – ferrite cores, magnets – can be detected and classififed

Special lighting system has been designed Detection of irregularities:

Irregularities in general – dynamic thresholding Compact irregularities – morphological method Elongated irregularities – morphological method General decision on quality of the tested object

Classification of irregularities: Training by showing examples

Segmentation and measurements Detailed, quantitative final decision

Project www site: http://www.tpo.org.pl/squash

Page 26: Structural Image Analysis in Investigation of Concrete Warsaw, October 21-23, 2002 Classification of defects on the surface of black ceramics Leszek Chmielewski,

26/25SIAIC 2002, Warsaw L. Chmielewski (PSWiP) et al., Classification of defects on the surface of black ceramics

ReferencesA. Jóźwik, L. Chmielewski, M. Skłodowski and W. Cudny. A proposition of the new feature space and its use

to construction of a fast minimum distance classifier. In Proc. 2nd Polish Conference on Computer Pattern Recognition Systems KOSYR 2001, pages 381-386, Miłków, Poland, May 28-31, 2001.

M. Nieniewski, L. Chmielewski, A. Jóźwik and M. Skłodowski. Morphological detection and feature-based classification of cracked regions in ferrites. MG&V, 8(4):699-712, 1999.

A. Jóźwik, L. Chmielewski, M. Skłodowski and W. Cudny. Class overlap rate as a design criterion for parallel Nearest Neighbour classifier. In Proc. 1st Polish Conference on Computer Pattern Recognition Systems KOSYR'99 Trzebieszowice, Poland, May 24-27, 1999.

A. Jóźwik, L. Chmielewski, M. Skłodowski and W. Cudny. A parallel net of (1-NN, k-NN) classifiers for optical inspection of surface defects in ferrites. MG&V, 7(1-2):99-112, 1998.

G. Vernazza, M. Lugg, T. Postupolski, A. Jóźwik, L. Chmielewski, D. Chetverikov and M. Peri. SQUASH: Standard Compliant Quality Control System for High-Level Ceramic Material Manufacturing. In Proc. INCO-COPERNICUS-INTAS Workshop on Advanced Ceramics and Alloys, pages 35-40, Brussels, Belgium, Mar 12-13, 1998. European Commission, Directoriate Generale XII.

L. Chmielewski, M. Skłodowski, W. Cudny, M. Nieniewski and A. Jóźwik. Optical system for detection and classification of surface defects in ferrites. In Proc. 3rd Symp. Image Processing Techniques (TPO'97), pages 1-13, Serock, Poland, Oct 29-31, 1997. Oficyna Wydawnicza Politechniki Warszawskiej.

M. Mari, C. Dambra, D. Chetverikov, J. Verestoy, A. Jóźwik, M. Nieniewski, M. Skłodowski, L. Chmielewski, W. Cudny and M. Lugg. The CRASH Project: Defect Detection and Classification in Ferrite Cores. In A. Del Bimbo, editor, Proc. 9th Int. Conf. Image Analysis and Processing, number 1310 in Lecture Notes in Computer Science, pages 781-787 (vol. II), Florence, Italy, Sept 17-19, 1997. Springer Verlag, Berlin.

A. Jóźwik, L. Chmielewski, W. Cudny and M. Skłodowski. A 1-NN preclassifier for fuzzy k-NN rule. In Proc. 13th Int. Conf. Pattern Recognition, pages D-234 - D-238, Wien, Austria, Aug 25-29, 1996. IAPR, Technical Univ. Vienna.

[Law80] K. I. Laws, Textured image segmentation, Univ. of Southern California, Image Processing Institute, USCIPI Report 940, Jan 1980

[Pra91] W. K. Pratt, Digital Image Processing, John Wiley, New York 1991. [WuCh92] C-M. Wu, Y-C. Chen, Statistical feature matrix for texture analysis, CVGIP: Graphical Models and

Image Processing, 54, 5, 1992, 407-419. [YBF95] G. Z. Yang, P. Burger, D. N. Firmin, S. R. Underwood, Structure Adaptive Anisotropic Filtering for

Magnetic Resonance Image Enhancement, Proc. 6th Int. Conf. CAIP, Prague, Czech Republic, Sept. 6-8, 1995, 384-391. Lecture Notes on Computer Science. Springer Verlag, 1995.