A method for microstructure similarity clustering and feature reconstruction for weathered weak muddy intercalations
- 131 Downloads
Weak muddy intercalations (WMI) are a type of geo-material with highly unstable mechanical properties, and thus they pose a great threat to the stability of rock slopes and in other rock engineering situations. In this paper, microstructure similarity-based clustering together with image fusion and reconstruction are used to study the microstructures of shallow weathered WMI. The study aims to obtain a reconstructed image of microstructure features that can represent a region to provide the basis for subsequent studies on WMI mechanical properties. The similarity of each microscopic WMI image is calculated using a similarity calculation model based on the microstructure parameters, and images are clustered based on their similarities. Then, image fusion technology is used to combine images in the same cluster. The results are as follows: (1) Similarity corresponding to a cumulative distribution probability of 80% is used as the clustering threshold; (2) The fused WMI microstructure image can represent the microstructure of a layer in the sample. In view of these findings, WMI microstructure clustering and feature reconstruction can provide evidence for studies on WMI lamina structures and failures involving these, which formed the basis for the assessment of the stability of slopes and other situations in which WMI are present.
KeywordsMuddy weak intercalations Microstructure Similarity calculation model Clustering Image fusion
The work was supported by the National Natural Science Foundation of China (Nos. 51574201) and the State Key Laboratory of Geo-hazard Prevention and Geo-environment Protection (Chengdu University of Technology) (SKLGP2015K006). Additional support was provided by the Scientific and Technical Youth Innovation Group (Southwest Petroleum University) (2015CXTD05).
Compliance with ethical standards
Conflict of interest
The authors declare that there is no conflict of interest regarding the publication of this paper.
- Cotecchia F, Cafaro F, Guglielmi S (2016) Microstructural changes in clays generated by compression explored by means of SEM and image processing. Procedia Engineer 158:57–62Google Scholar
- Deza MM, Deza E (2009) Encyclopedia of distances. Springer, Berlin, p 1–583Google Scholar
- Jiang C, Zhou X, Tao G, Chen D (2016) Experimental study on the performance and micro-structure of cementitious materials made with dune sand. Adv Mater Sci Eng. https://doi.org/10.1155/2016/2158706
- Kim SK, Kang ST, Kim JK, Jang IY (2017) Effects of particle size and cement replacement of LCD glass powder in concrete. Adv Mater Sci Eng. https://doi.org/10.1155/2017/3928047
- Lin S-z, Wang D-j, Wang X-x, Zhu X-h (2016) Multi-band texture image fusion based on the embedded multi-scale decomposition andpossibility theory. Spectrosc Spectr Anal 36(7):2337–2343Google Scholar
- Ng AY, Jordan MI, Weiss Y (2002) On spectral clustering: analysis and an algorithm. In: Jordan MI, LeCun Y, Solla SA (eds) Advances in neural information processing systems. MIT Press, Cambridge, pp 849–856Google Scholar
- Wang C-l, Wang H-W, Hu B-l, Wen J, Xu J, Li X-J (2016) A novel spatial-spectral sparse representation for hyperspectral image classification based on neighborhood segmentation. Spectrosc Spectr Anal 36(9):2919–2924Google Scholar
- Zheng L-N (2012) Reaserch of failure mechanism and the local failure zones for consequent slope based on strain softening theory. Southwest Jiao Tong University, ChengduGoogle Scholar
- Zheng Y, Liu J, Hu Q, Cai Q (2014) Study on micro-structure of muddy intercalation using SEM method. Electron J Geotech Eng 19:9953–9963Google Scholar