Hostname: page-component-cd9895bd7-8ctnn Total loading time: 0 Render date: 2024-12-19T11:55:10.320Z Has data issue: false hasContentIssue false
  • English
  • Français

Phase Classification by Mean Shift Clustering of Multispectral Materials Images

Published online by Cambridge University Press:  26 June 2013

Diego Schmaedech Martins ,
Victor M. Galván Josa ,
Gustavo Castellano  and
José A.T. Borges da Costa
Diego Schmaedech Martins
Affiliation:
Programa de Pós-Graduação em Informática, Universidade Federal de Santa Maria, 97105-900 Santa Maria, RS, Brazil
Victor M. Galván Josa
Affiliation:
FaMAF, Universidad Nacional de Córdoba, Medina Allende s/n, Ciudad Universitaria, 5000 Córdoba, Argentina IFEG-CONICET, Medina Allende s/n, Ciudad Universitaria, 5000 Córdoba, Argentina
Gustavo Castellano
Affiliation:
FaMAF, Universidad Nacional de Córdoba, Medina Allende s/n, Ciudad Universitaria, 5000 Córdoba, Argentina IFEG-CONICET, Medina Allende s/n, Ciudad Universitaria, 5000 Córdoba, Argentina
José A.T. Borges da Costa*
Affiliation:
Departamento de Física, Universidade Federal de Santa Maria, 97105-900 Santa Maria, RS, Brazil
*
*Corresponding author. E-mail: [email protected]
Get access

Abstract

A mean-shift clustering (MSC) algorithm is introduced as a valuable alternative to perform materials phase classification from multispectral images. As opposed to other multivariate statistical techniques, such as factor analysis or principal component analysis (PCA), clustering techniques directly assign a class label to each pixel, so that their outputs are phase segmented images, i.e., there is no need for an additional segmentation algorithm. On the other hand, as compared to other clustering procedures and classification methods, such as segmentation by thresholding of multiple spectral components, MSC has the advantages of not requiring previous knowledge of the number of data clusters and not assuming any shape for these clusters, i.e., neither the number nor the composition of the phases must be previously known. This makes MSC a particularly useful tool for exploratory research, assisting phase identification of unknown samples. Visualization and interpretation of the results are also simplified, since the information content of the output image does not depend on the particular choice of the content of the color channels. We applied MSC to the analysis of two sets of X-ray maps acquired in scanning electron microscopes equipped with energy-dispersive detection systems. Our results indicate that MSC is capable of detecting additional phases, not clearly identified through PCA or multiple thresholding, with a very low empirical reject rate.

Keywords

phase classificationX-ray mapsSEMimage segmentationmean-shift clustering
Type
Techniques and Software Development
Information
Microscopy and Microanalysis , Volume 19 , Issue 5 , October 2013 , pp. 1266 - 1275
Copyright
Copyright © Microscopy Society of America 2013 

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

Allen, J.L., Johnson, C.L., Heumann, M.J., Gooley, J. & Gallin, W. (2012). New technology and methodology for assessing sandstone composition: A preliminary case study using a quantitative electron microscope scanner (QEMScan). In Mineralogical and Geochemical Approaches to Provenance: Geological Society of America Special Paper, Rasbury, E.T., Hemming, S.R. & Riggs, N.R. (Eds.), vol. 487, 177–194. Boulder, CO: The Geological Society of America.Google Scholar
Bentz, D.P., Stutzman, P.E., Haecker, C.J. & Remond, S. (1999). SEM/X-ray imaging of cement-based materials. In Proceedings of the 7th Euroseminar on Microscopy Applied to Building Materials, Pietersen, H.S., Larbi, J.A. & Janssen, H.H.A. (Eds.), pp. 457–466. Delft, the Netherlands: Delft University of Technology.Google Scholar
Bertolino, S.R. & Fabra, M. (2003). Provenance and ceramic technology of pot sherds from ancient Andean cultures at the Ambato valley, Argentina. Appl Clay Sci 24, 21–34.CrossRefGoogle Scholar
Bertolino, S.R., Galván, V., Carreras, A., Laguens, A., De La Fuente, G. & Riveros, A. (2009). X-ray techniques applied to surface painting. X-Ray Spectrom 38, 95–102.Google Scholar
Bo, S., Ding, L., Li, H., Di, F. & Zhu, C. (2009). Mean shift-based clustering analysis of multispectral remote sensing imagery. Int J Rem Sen 30(4), 817–827.CrossRefGoogle Scholar
Borges da Costa, J.A.T., Rosa, M., Takehara, L., Ornellas, M.C. & Vasconcellos, M.A.Z. (2007). Grain segmentation of hematite-rich ore by multivariate analysis of polarized light image stacks. In VIII Simpósio Brasileiro de Minério de Ferro. Salvadore, Bahia, Brazil, September 18–27, 2007. São Paulo, SP: Associação Brasileira de Metalurgia e Materiais.Google Scholar
Carling, G.T., Fernandez, D.P. & Johnson, W.P. (2012). Dust-mediated loading of trace and major elements to Wasatch Mountain snowpack. Sci Total Environ 432, 67–77.CrossRefGoogle ScholarPubMed
Cellier, F., Oriot, H. & Nicolas, J.M. (2005). Introduction of the mean shift algorithm in SAR imagery: Application to shadow extraction for building reconstruction. In Proceedings of the Earsel 3D Remote Sensing Workshop, Porto, Portugal, June 10–11, 2005. Google Scholar
Cheng, Y. (1995). Mean shift, mode seeking, and clustering. IEEE Trans Pattern Anal Machine Intell 17(8), 790–799.CrossRefGoogle Scholar
Comaniciu, D. (2003). An algorithm for data-driven bandwidth selection. IEEE Trans Patt Anal Mach Intell 25(2), 281–288.Google Scholar
Comaniciu, D. & Meer, P. (1997). Robust analysis of feature spaces: Color image segmentation. IEEE Proceedings Conf Computer Vision Pattern Recognition, pp. 750–755. Google Scholar
Comaniciu, D. & Meer, P. (2002). Mean shift: A robust approach toward feature space analysis. IEEE Trans Pattern Anal Machine Intell 24, 603–619.Google Scholar
Comaniciu, F., Ramesh, V. & Meer, P. (2001). The variable bandwidth mean shift and data-driven scale selection. In Proceedings of the Eighth International Conference on Computer Vision, vol. 1, pp. 438–445. Washington, DC: IEEE.Google Scholar
Ding, Q. & Colpan, M. (2006). Decision tree induction on hyper-spectral cement images. Int J Inform Math Sci 2(3), 169–175.Google Scholar
Finkston, B. (2006). Mean-shift clustering. Available at http://www.mathworks.com (accessed October, 19 2010).Google Scholar
Fukunaga, K. & Hostetler, L.D. (1975). The estimation of the gradient of a density function, with applications in pattern recognition. IEEE Trans Info Theory 21(1), 32–40.Google Scholar
Galván Josa, V., Bertolino, S., De La Fuente, G., Riveros, J. & Castellano, G. (2009). Caracterización de Pinturas Arqueológicas Mediante el Procesamiento de Mapas de Rayos X. Acta Microscopica 18(Suppl C), 385–386.Google Scholar
Gernet, U. (2008). Comparing the Si(Li)-detector and the silicon drift detector (SDD) using EDX in SEM. In 14th European Microscopy Congress, Luysberg, M., Tillmann, K. & Weirich, T. (Eds.), pp. 697–698. Aachen, Germany: Springer Berlin Heidelberg.Google Scholar
Götze, J., Plötze, M., Götte, T., Neuser, R.D. & Richter, D.K. (1997). Cathodoluminescence (CL) and electron paramagnetic resonance (EPR) studies of clay minerals. Mineral Petr 76, 195–212.Google Scholar
Harris, R.J. (2001). A Primer of Multivariate Statistics, 3rd ed., pp. 40–41. Mahwah, NJ: Lawrence Erlbaum Publishers.Google Scholar
Humphreys, F.J. (2001). Grain and subgrain characterisation by electron backscatter diffraction. J Mater Sci 36, 3833–3854.Google Scholar
Jain, A.K. (2010). Data clustering: 50 years beyond K-means. Patt Recog Lett 31, 651–666.Google Scholar
Keenan, M.R. (2008). Exploiting spatial-domain simplicity in spectral image analysis. Surf Interface Anal 41, 79–87.Google Scholar
Kotula, P.G. (2002). Spectral imaging: Towards quantitative X-ray microanalysis. Microsc Microanal 8(Suppl 2), 440–441.Google Scholar
Kotula, P.G. & Keenan, M.R. (2006). Application of multivariate statistical analysis to STEM X-ray spectral images: Interfacial analysis in microelectronics. Microsc Microanal 12, 538–544.Google Scholar
Kotula, P.G., Keenan, M.R. & Michael, J.R. (2003). Automated analysis of SEM X-ray spectral images: A powerful new microanalysis tool. Microsc Microanal 9, 1–17.Google Scholar
Lizarazo, I. (2008). SVM-based segmentation and classification of remotely sensed data. Int J Remote Sens 29(24), 7277–7283.CrossRefGoogle Scholar
MacRae, C.M., Wilson, N.C. & Brugger, J. (2009). Quantitative cathodoluminescence mapping with application to a Kalgoorlie scheelite. Microsc Microanal 15, 222–230.CrossRefGoogle ScholarPubMed
MacRae, C.M., Wilson, N.C., Johnson, S.A., Phillips, P.L. & Otsuki, M. (2005). Hyperspectral mapping combining cathodoluminescence and X-ray collection in an electron microprobe. Microsc Res Tech 67(5), 271–277.Google Scholar
Miranda, A.N., Pinheiro, D.P. & Borges da Costa, J.A.T. (2004). QuantiPhase, Informtica, Universidade Federal de Santa Maria, Brazil. Available at http://www-usr.inf.ufsm.br/~miranda/QuantiPhase.php (accessed March 1, 2013).Google Scholar
Neal, B. & Russ, J. (2004). Principal components analysis of multispectral image data. Microsc Today 12, 36–38.Google Scholar
Pirard, E. (2004). Multispectral imaging of ore minerals in optical microscopy. Mineral Mag 68(2), 323–333.Google Scholar
Pirard, E. & Lebichot, S. (2005). Automated identification of iron oxides under the optical microscope. Mineral Georesources and Geo-Imaging Group, GeomaC Department Belgium. Google Scholar
Pirard, E., Lebichot, S. & Krier, W. (2007). Particle texture analysis using polarized light imaging and gray level intercepts. Int J Miner Process 84, 299–309.Google Scholar
Russ, J. (2006). The Image Processing Handbook, 5th ed. Boca Raton, FL: CRC Press.Google Scholar
Shapiro, L. & Stockman, G. (2001). Computer Vision. Englewood Cliffs, NJ: Prentice Hall.Google Scholar
Stork, C.L. & Keenan, M. (2010). Advantages of clustering in the phase classification of hyperspectral materials image. Microsc Microanal 47, 1–17.Google Scholar