Preprints
https://doi.org/10.48550/arXiv.2103.12821
https://doi.org/10.48550/arXiv.2103.12821
 
01 Jun 2022
01 Jun 2022

Detecting micro fractures: A comprehensive comparison of conventional and machine-learning based segmentation methods

Dongwon Lee1, Nikolaos Karadimitriou1, Matthias Ruf1, and Holger Steeb1,2 Dongwon Lee et al.
  • 1Institute of Applied Mechanics (CE), Pfaffenwaldring 7, University of Stuttgart,70569, Germany
  • 2Stuttgart Center for Simulation Science, Pfaffenwaldring 5a, University of Stuttgart,70569, Germany

Abstract. Studying porous rocks with X-Ray Computed Tomography (XRCT) has been established as a standard procedure for the non-destructive characterization of flow and transport in porous media. Despite the recent advances in the field of XRCT, various challenges still remain due to the inherent noise and imaging artefacts in the produced data. These issues become even more profound when the objective is the identification of fractures, and/or fracture networks. One challenge is the limited contrast between the regions of interest and the neighboring areas, which can mostly be attributed to the minute aperture of the fractures. In order to overcome this challenge, it has been a common approach to apply various digital image processing steps, such as filtering, to enhance the signal-to-noise ratio. Additionally, segmentation methods based on threshold/morphology schemes have been employed to obtain enhanced information from the features of interest. However, this workflow needs a skillful operator to fine-tune its input parameters, and the required computation time significantly increases due to the complexity of the available methods, and the large volume of an XRCT data-set. In this study, based on a data-set produced by the successful visualization of a fracture network in Carrara marble with μXRCT, we present the results from five segmentation methods, three conventional and two machine learning-based ones. The objective is to provide the interested reader with a comprehensive comparison between existing approaches, while presenting the operating principles, advantages and limitations, to serve as a guide towards an individualized segmentation workflow. The segmentation results from all five methods are compared to each other in terms of quality and time efficiency. Due to memory limitations, and in order to accomplish a fair comparison, all the methods are employed in a 2D scheme. The output of the 2D U-net model, which is one of the adopted machine learning-based segmentation methods, shows the best performance regarding the quality of segmentation and the required processing time.

Journal article(s) based on this preprint

27 Sep 2022
Detecting micro fractures: a comprehensive comparison of conventional and machine-learning-based segmentation methods
Dongwon Lee, Nikolaos Karadimitriou, Matthias Ruf, and Holger Steeb
Solid Earth, 13, 1475–1494, https://doi.org/10.5194/se-13-1475-2022,https://doi.org/10.5194/se-13-1475-2022, 2022
Short summary

Dongwon Lee et al.

Interactive discussion

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on egusphere-2022-400', Anonymous Referee #1, 03 Jul 2022
    • AC1: 'Reply on RC1', Dongwon Lee, 08 Aug 2022
  • RC2: 'Comment on egusphere-2022-400', Anonymous Referee #2, 05 Jul 2022
    • AC2: 'Reply on RC2', Dongwon Lee, 08 Aug 2022

Interactive discussion

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on egusphere-2022-400', Anonymous Referee #1, 03 Jul 2022
    • AC1: 'Reply on RC1', Dongwon Lee, 08 Aug 2022
  • RC2: 'Comment on egusphere-2022-400', Anonymous Referee #2, 05 Jul 2022
    • AC2: 'Reply on RC2', Dongwon Lee, 08 Aug 2022

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision
AR by Dongwon Lee on behalf of the Authors (08 Aug 2022)  Author's response    Author's tracked changes    Manuscript
ED: Publish as is (15 Aug 2022) by David Healy
ED: Publish as is (15 Aug 2022) by Federico Rossetti(Executive Editor)

Journal article(s) based on this preprint

27 Sep 2022
Detecting micro fractures: a comprehensive comparison of conventional and machine-learning-based segmentation methods
Dongwon Lee, Nikolaos Karadimitriou, Matthias Ruf, and Holger Steeb
Solid Earth, 13, 1475–1494, https://doi.org/10.5194/se-13-1475-2022,https://doi.org/10.5194/se-13-1475-2022, 2022
Short summary

Dongwon Lee et al.

Data sets

micro-XRCT data set of Carrara marble with artificially created crack network: fast cooling down from 600°C Matthias Ruf;Holger Steeb https://doi.org/10.18419/darus-682

Model code and software

Fracture network segmentation Dongwon Lee;Nikolaos Karadimitriou;Holger Steeb https://doi.org/10.18419/darus-1847

Dongwon Lee et al.

Viewed

The metrics are limited to the final revised paper since the preprint was posted outside of Copernicus Publications.

Total article views: 177 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
177 0 0 177 2 2
  • HTML: 177
  • PDF: 0
  • XML: 0
  • Total: 177
  • BibTeX: 2
  • EndNote: 2
Views and downloads (calculated since 01 Jun 2022)
Cumulative views and downloads (calculated since 01 Jun 2022)

Viewed (geographical distribution)

The metrics are limited to the final revised paper since the preprint was posted outside of Copernicus Publications.

Total article views: 147 (including HTML, PDF, and XML) Thereof 147 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 27 Sep 2022
Download

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Short summary
This research article focuses on filtering and segmentation methods employed in high resolution µXRCT studies for crystalline rocks, bearing fractures, or fracture networks, of very small aperture. Specifically, we focus on the identification of artificially induced, via quenching, fractures in Carrara marble samples. Results from the same data-set from all adopted five different methods were produced and compared with each other in terms of their output quality and time efficiency.