Accessible Unlicensed Requires Authentication Published by De Gruyter April 29, 2020

Application of Trainable Segmentation to Microstructural Images Using Low-alloy Steels as an Example

Anwendung trainierbarer Segmentierungen auf Gefügeaufnahmen am Beispiel niedriglegierter Stähle
M. Müller, D. Britz and F. Mücklich
From the journal Practical Metallography

Abstract

One of the most important but yet partly neglected steps in quantitative microstructural analysis is microscopic image segmentation. Despite major advances in the currently available segmentation methods and first of all, machine learning segmentation, the metal industry in many cases still relies on segmentation by thresholding. Based on microstructural images of low-alloy steels, this work will show to what extent machine learning segmentation can accomplish a significant quality improvement compared to the traditional thresholding method.

Kurzfassung

Einer der wichtigsten, aber dennoch teilweise vernachlässigten Schritte in der quantitativen Gefügeanalyse ist die Segmentierung der Mikroskop-Aufnahme. Trotz großer Fortschritte in den zur Verfügung stehenden Segmentierungstechniken, in erster Linie Segmentierungen mit Hilfe von maschinellem Lernen, wird in der Metallbranche immer noch oft die Schwellwert-Segmentierung eingesetzt. In dieser Arbeit soll am Beispiel von Gefügeaufnahmen niedriglegierter Stähle gezeigt werden, inwiefern Segmentierungen mittels maschinellem Lernen die Qualität gegenüber der klassischen Schwellwert-Segmentierung signifikant verbessern können.


Translation: V. Müller


References / Literatur

[1] Jähne, B.: Digital Image Processing, 5th ed., Springer Berlin Heidelberg, 2002. 10.1007/978-3-662-04781-1Search in Google Scholar

[2] Sahoo, P.K.; Soltani, S.; Wong, A.K.C.: Comput. Vision, Graph. Image Process.41 (1988) 233260. 10.1016/0734-189X(88)90022-9Search in Google Scholar

[3] Otsu, N.: IEEE Trans. Syst. Man. Cybern.9 (1979) 6266. 10.1109/TSMC.1979.4310076Search in Google Scholar

[4] Senthilkumaran, N.; Rajesh, R.: in: 2009 Int. Conf. Adv. Recent Technol. Commun. Comput., IEEE, 2009, pp. 844846.Search in Google Scholar

[5] WenshuoGao; XiaoguangZhang; LeiYang; HuizhongLiu: in: 2010 3rd Int. Conf. Comput. Sci. Inf. Technol., IEEE, 2010, pp. 6771.Search in Google Scholar

[6] Canny, J.: IEEE Trans. Pattern Anal. Mach. Intell.PAMI–8 (1986) 679698. 10.1109/TPAMI.1986.4767851Search in Google Scholar

[7] Illingworth, J.; Kittler, J.: Comput. Vision, Graph. Image Process.44 (1988) 87116. 10.1016/S0734-189X(88)80033-1Search in Google Scholar

[8] Jain, A.K.: Fundamentals of Digital Image Processing, Prentice Hall, 1989.Search in Google Scholar

[9] Feng-Ping, A.; Zhi-Wen, L.: Biomed. Signal Process. Control53 (2019) 101589. 10.1016/j.bspc.2019.101589Search in Google Scholar

[10] Chai, Y.; Liu, H.; Xu, J.: Appl. Soft Comput. (2019) 105890. 10.1016/j.asoc.2019.105890Search in Google Scholar

[11] Karimi, D.; Zeng, Q.; Mathur, P.; Avinash, A.; Mahdavi, S.; Spadinger, I.; Abolmaesumi, P.; Salcudean, S.E.: Med. Image Anal.57 (2019) 186196. 10.1016/j.media.2019.07.005Search in Google Scholar

[12] Hesamian, M.H.; Jia, W.; He, X.; Kennedy, P.: J. Digit. Imaging3 (2019) 582–596. 10.1007/s10278-019-00227-xSearch in Google Scholar

[13] Wang, S.; Cao, G.; Wei, B.; Yin, Y.; Yang, G.; Li, C.: Biomed. Eng. Online12 (2013) 114. 10.1186/1475-925X-12-1Search in Google Scholar

[14] Burget, R.; Uher, V.; Masek, J.: Trainable Segmentation Based on Local-Level and Segment-Level Feature Extraction, 2012.Search in Google Scholar

[15] Komenda, J.: Mater. Charact.46 (2001) 8792. 10.1016/S1044-5803(01)00106-1Search in Google Scholar

[16] Papa, J.P.; Pereira, C.R.; De Albuquerque, V.H.C.; Silva, C.C.; Falcão, A.X.; Tavares, J.M.R.S.: Lect. Notes Comput. Sci. (Including Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics) 6636 LNCS (2011) 456468.Search in Google Scholar

[17] Bulgarevich, D.S.; Tsukamoto, S.; Kasuya, T.; Demura, M.; Watanabe, M.: Sci. Rep.8 (2018) 39. 10.1038/s41598-018-20438-6Search in Google Scholar

[18] De Albuquerque, V.H.C.; Cortez, P.C.; De Alexandria, A.R.; Tavares, J.M.R.S.: Nondestruct. Test. Eval.23 (2008) 273283. 10.1080/10589750802258986Search in Google Scholar

[19] De Albuquerque, V.H.C.; Silva, C.C.; Menezes, T.I. de S.; Farias, J.P.; Tavares, J.M.R.S.: Microsc. Res. Tech.74 (2011) 3646. 10.1002/jemt.20870Search in Google Scholar

[20] Azimi, S.M.; Britz, D.; Engstler, M.; Fritz, M.; Mücklich, F.: Sci. Rep.8 (2018) 114. 10.1038/s41598-018-20037-5Search in Google Scholar

[21] DeCost, B.L.; Lei, B.; Francis, T.; Holm, E.A.: Microsc. Microanal.25 (2019) 2129. 10.1017/S1431927618015635Search in Google Scholar

[22] Britz, D.; Hegetschweiler, A.; Roberts, M.; Mücklich, F.: Mater. Perform. Charact.5 (2016) 553563. 10.1520/MPC20160067Search in Google Scholar

[23] Britz, D.; Steimer, Y.; Mücklich, F.: in: 100 Years E04 Dev. Metallogr. Stand., ASTM International, 2019, pp. 130151. 10.1520/STP160720170242Search in Google Scholar

[24] Zeiss Intellesis, (n.d.). (accessed November 15, 2019).Search in Google Scholar

[25] Andrew, M.: Comput. Geosci.22 (2018) 15031512. 10.1007/s10596-018-9768-ySearch in Google Scholar

[26] Dragonfly, (n.d.). (accessed November 15, 2019).Search in Google Scholar

[27] Arganda-Carreras, I.; Kaynig, V.; Rueden, C.; Eliceiri, K.W.; Schindelin, J.; Cardona, A.; Seung, H.S.: Bioinformatics33 (2017) 24242426. 10.1093/bioinformatics/btx180Search in Google Scholar

[28] Trainable Weka Segmentation, (n.d.). (accessed November 15, 2019).Search in Google Scholar

[29] ImageJ, (n.d.). (accessed November 15, 2019).Search in Google Scholar

[30] Richter, S.: Statistisches Und Maschinelles Lernen, Springer Berlin Heidelberg, 2019. 10.1007/978-3-662-59354-7Search in Google Scholar

[31] Kotthoff, L.; Thornton, C.; Hutter, F.: (accessed November 15, 2019).Search in Google Scholar

[32] Britz, D.; Webel, J.; Gola, J.: Pract. Metallogr.54 (2017) 685696. 10.3139/147.110484Search in Google Scholar

[33] Camilus, K.S.; Govindan, V.K.: I.J. Image, Graph. Signal Process.5 (2012) 113. 10.5815/ijigsp.2012.05.01Search in Google Scholar

[34] Graph Cut – Image J, (n.d.). (accessed November 18, 2019).Search in Google Scholar

[35] Boykov, Y.; Kolmogorov, V.: IEEE Transactions on PAMI26 (2004) 11241137. 10.1109/TPAMI.2004.60Search in Google Scholar

Received: 2020-02-04
Accepted: 2020-02-07
Published Online: 2020-04-29
Published in Print: 2020-05-15

© 2020, Carl Hanser Verlag, München