The Use of Texture Analysis in Optical Inspection of Manufacturing Processes

Abstract:

The paper presents a method of optical inspection of manufacturing processes in the visible band. The presented model of the test bench was based on a monochrome CCD camera that allows one to conduct research in the field of digital image analysis. The wavelet transform was proposed for the detection of objects. This wavelet transform allows examining pictures with use of textures. The article presents an analysis of the results of experiments for selected objects using the proposed method, allowing the evaluation of the correctness of classification. In the article, the advantages of using the proposed method in a sample manufacturing process to recognizing surface defects (e.g. cracks or scratches) is discussed.

• Periodical: Solid State Phenomena (Volume 237)
• Pages: 95-100
• DOI: https://doi.org/10.4028/www.scientific.net/SSP.237.95
• Online since: August 2015
• Authors: Tomasz Marciniak*, Mirosław Maszewski, Beata Marciniak, Zbigniew Lutowski
• Keywords: Optical Inspection, Texture Analysis, Wavelet Transform

References

[1] Aleixos N., Blasco J., Navarrón F., Molto E.: Multispectral inspection of citrus in real-time using machine vision and digital signal processors, Computers and Electronics in Agriculture 2002, vol. 33, pp.121-137.

DOI: 10.1016/s0168-1699(02)00002-9

[2] Boroński D., Giesko T., Marciniak T., Lutowski Z., Bujnowski S.: The detection and measurement of fatigue crack length with the use of the FatigueVIEW system, Przegląd Mechaniczny 4 (2014), pp.21-27.

DOI: 10.4028/www.scientific.net/kem.598.26

[3] Choraś R.S.: Komputerowa wizja, Metody interpretacji i identyfikacji obiektów, Akademicka oficyna wydawnicza EXIT, Warszawa 2005, ISBN 83-87674-89-3.

[4] Grossman A., Morlet J.: Decomposition of Hardy functions into square integrable wavelets of constant shape, CBMS-NSF Series on Applied Mathematics, Society for Industrial and Applied Mathematics, (1984).

DOI: 10.1515/9781400827268.126

[5] Lutowski Z., Marciniak T., Bujnowski S., Boroński D.: The optical system for elongation measurement of highly deformable materials, Maintenance problems 2-(2013).

[6] Mallat S.: A Theory for Multiresolution Signal Decomposition: The Wavelet Representation, IEEE Transasctions on Pattern Analysies and Machine Inteligence, IEEE (1989).

DOI: 10.1109/34.192463

[7] Marciniak T., Lutowski Z., Bujnowski S., Boroński D., Czajka P.: Dual-Band Experimental System For subsurface Cracks Testing, Materials Science Forum, Fatigue Failure and Fracture Mechanics, Trans Tech Publications 2012, vol. 726, pp.222-226.

DOI: 10.4028/www.scientific.net/msf.726.222

[8] Marciniak B., Marciniak T., Lutowski Z., Bujnowski S.: Usage of digital image correlation in analysis of cracking, Image Processing and Communications 2012, vol. 17, no. 3, pp.21-29.

DOI: 10.2478/v10248-012-0019-x

[9] Minkina W.: Pomiary termowizyjne – przyrządy i metody, Wydawnictwo Politechniki Częstochowskiej, Częstochowa (2004).

[10] Oliferuk W.: Termografia podczerwieni w nieniszczących badaniach materiałów i urządzeń, Biuro Gamma, Warszawa (2008).

[11] Oppenheim A.V., Sheafer R.W.: Digital Signal Processing, Signal Processing Series, Pretince Hall (1979).

[12] Szczepanik M., Stabik J., Wróbel G., Wierzbicki Ł.: Wykorzystanie systemów termowizyjnych do badań materiałów polimerowych, Modelowanie inżynierskie, Gliwice 2008, pp.279-286.

[13] Unay D., Gosselin B., Kleynen O., Leemans W., Destain M., Debeir O.: Automatic grading of Bi-colored apples by multispectral machine vision, Computers and Electronics in Agriculture 2011, vol. 75, pp.204-212.

DOI: 10.1016/j.compag.2010.11.006

[14] Domek S., Grudziński M., Okarma K., Pajor M.: Korekcja nieliniowości charakterystyki projekcji światła strukturalnego w wizyjnym systemie pozycjonowania przedmiotu obrabianego. Przegląd Elektrotechniczny, R. 88 nr 10a, str. 143-146, (2012).