ABSTRACT
Multiphase materials are encountered in several areas of science and technology. Their properties are determined by the fraction of the phases (material compounds) constituting the composite material. Therefore, the quantitative characterization of phase fractions is highly demanded and has been the subject of extensive studies. To this end, a widely used technique is the segmentation of top-down back-scattered electron SEM (BSE-SEM) images given that different phases are depicted with pixel collections of different luminosity. Gaussian mixture models (GMM) are one the most popular and easily implemented methods to segment the BSE-SEM images through the deconvolution of their histograms. However, the accuracy and the limitations of their application have not been fully investigated. The aim of this paper is to design a neural-network approach to fill this gap and provide a fast tool for the automatic evaluation of the accuracy of GMM predictions for all material phases based on the inspection of the measured SEM image histogram alone. The proposed tool facilitates the decision-making process of an SEM user concerning the optimum choice of a segmentation method.
- Su, Z., Decencière, E., Nguyen, TT. Artificial neural network approach for multiphase segmentation of battery electrode nano-CT images. npj Comput Mater 8, 30 (2022). https://doi.org/10.1038/s41524-022-00709-7Google Scholar
- Daya R. Nhuchhen, Song P. Sit, David B. Layzell, 2022, Decarbonization of cement production in a hydrogen economy, Applied Energy 317, https://doi.org/10.1016/j.apenergy.2022.119180.Google Scholar
- Koumpouri Dimitra, Karatasios Ioannis, Psycharis Vassilis, Giannakopoulos Ioannis, Katsiotis Marios, Kilikoglou Vassilis ,2021, Effect of clinkering conditions on phase evolution and microstructure of Belite Calcium-Sulpho-Aluminate cement clinker, Cement and Concrete Research 147, https://doi.org/10.1016/j.cemconres.2021.106529Google Scholar
- Gonzaga Carla, Okada Cristina, Cesar Paulo, Miranda Walter, Yoshimura Humberto, 2009, Effect of processing induced particle alignment on the fracture toughness and fracture behavior of multiphase dental ceramics, Dental Materials 25, https://doi.org/10.1016/j.dental.2009.03.013.Google Scholar
- Christiane Rossler, Dominik Zimmer, Patrick Trimby, Horst-Michael Ludwig, 2022, Chemical crystallographic characterization of cement clinkers by EBSD-EDS analysis in the SEM, Cement and Concrete Research 154, https://doi.org/10.1016/j.cemconres.2022.106721Google ScholarCross Ref
- Georget Fabien, Wilson William, Scrivener Karen, 2022, Simple automation of SEM-EDS spectral maps analysis with Python and the edxia framework, Journal of Microscopy 286, https://doi.org/10.1111/jmi.13099Google Scholar
- Hu Chuanlin, Ma Hongyan, 2016, Statistical analysis of backscattered electron image of hydrated cement paste, Advances in Cement Research 28, http://dx.doi.org/10.1680/jadcr.16.00002Google Scholar
- Chatzigeorgiou Manolis, Vrigkas Michalis, Boukos Nikos, Katsiotis Marios, Vassilios Constantoudis, [in writing] Gaussian Mixture model Segmentation on SEM Backscattered electron images of multiphase materials: Prospects and limitations, Journal of MicroscopyGoogle Scholar
- Chollet, F., & others. (2015). Keras. GitHub. Retrieved from https://github.com/fchollet/kerasGoogle Scholar
- Günter Klambauer, Thomas Unterthiner, Andreas Mayr, Sepp Hochreiter,2017, Self-Normalizing Neural Networks, https://doi.org/10.48550/arXiv.1706.02515Google Scholar
- Diederik P. Kingma, Jimmy Ba , 2014 , https://doi.org/10.48550/arXiv.1412.6980Google Scholar
Index Terms
- Machine Learning evaluation of microscopy image segmentation methods: The case of Gaussian Mixture Models
Recommendations
Brain MRI image segmentation based on learning local variational Gaussian mixture models
Measuring the distribution of major brain tissues, including the gray matter, white matter and cerebrospinal fluid (CSF), using magnetic resonance imaging (MRI) has attracted extensive research efforts. Many brain MRI image segmentation methods in the ...
Image Segmentation Using Correlative Histogram Modeled by Gaussian Mixture
ICDIP '09: Proceedings of the International Conference on Digital Image ProcessingIn this paper we address the problem of gray image segmentation. Our approach falls in category of histogram based thresholding methods. From image we first construct a correclative histogram, based on intensity of image pixels and the average intensity ...
A robust modified Gaussian mixture model with rough set for image segmentation
Accurate image segmentation is an essential step in image processing, where Gaussian mixture models with spatial constraint play an important role and have been proven effective for image segmentation. Nevertheless, most methods suffer from one or more ...
Comments