Detection of pore space in CT soil images using artificial neural networks

Cortina-Januchs, M. G.; Quintanilla-Domínguez, J.; Vega-Corona, A.; Tarquis, Ana M.; Andina, Diego

doi:10.5194/bg-8-279-2011

Cited by 40 publications

(11 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, the images might provide a weak contrast at the solid–void interface, in some cases creating a challenge to performing an appropriate segmentation and delimiting the pore space, as described by Cortina‐Januchs et al . ().…”

Section: Introductionmentioning

confidence: 97%

See 1 more Smart Citation

Pore detection in 3‐D CT soil samples through an improved sub‐segmentation method

Ojeda-Magaña

Domínguez²,

Ruelas

et al. 2018

European J Soil Science

Self Cite

View full text Add to dashboard Cite

Summary X‐ray computer tomography (CT) is a non‐invasive technique for image acquisition. Recent technological advances have enabled reliable and high‐resolution images to be obtained. In soil samples, for example, this subserves the identification of pores and their structure and the analysis of their geometric characteristics. However, the lack of contrast between pores and solids in soil samples makes it difficult to identify the pores, and it poses problems for their connectivity when a three‐dimensional (3‐D) reconstruction is made from a group of consecutive 2‐D images obtained with a scanner. To solve this problem, an improved sub‐segmentation method, which had been developed and tested previously, was applied in this research to achieve a better identification of the pore space and consequently the solid space in the 2‐D slices of the image, followed by a 3‐D reconstruction of the soil sample. In this study, two soil samples were used, one real soil sample with 255 2‐D CT consecutive images and a synthetic image with 215 2‐D images. The latter sample was used only to evaluate the robustness of the improved sub‐segmentation method and the results from analysis of the pore connectivity in a known structure. The results obtained with the improved sub‐segmentation were compared with those of traditional clustering algorithms for image segmentation by k‐means, fuzzy c‐means and Otsu's methods. The results were promising, and the 3‐D reconstruction presents a realistic structure for the continuity and coincidence of the shapes of the pores in the consecutive images. In addition, the pore regions detected have a small non‐uniformity (NU) value, which indicates both good pore detection and homogeneity, which facilitates pore connectivity between the different 2‐D images. Highlights Pore structure is not disturbed by the acquisition of images with computer tomography. Pore spaces were identified properly with the improved sub‐segmentation. Pore continuity by position, size and shape was observed through the 2‐D consecutive images. Realistic 3‐D reconstruction was feasible for the pore spaces in soil samples.

show abstract

Section: Introductionmentioning

confidence: 97%

“…In this study, we initially considered the results of Cortina‐Januchs et al . () and Ojeda‐Magaña et al . ().…”

Section: Introductionmentioning

confidence: 97%

Pore detection in 3‐D CT soil samples through an improved sub‐segmentation method

Ojeda-Magaña

Domínguez²,

Ruelas

et al. 2018

European J Soil Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…Iassonov et al (2009) have performed a wide review of different segmentation methods typically used in the field of geoscience. Other methods that appear to be promising for soil applications are the clustering and entropy-based methods (Cortina-Januchs et al, 2011;Iassonov and Tuller, 2010;Iassonov et al, 2009;Sezgin and Sankur, 2004), as well as the improvement that Houston made to the Indicator Kriging algorithm (Houston et al, 2013b) and the Schlüter method for determining a two-level intensity threshold (Schlüter et al, 2010). The fully automated segmentation method reported by Hapca et al (2013) also shows promise.…”

Section: Introductionmentioning

confidence: 99%

Identification of pore spaces in 3D CT soil images using PFCM partitional clustering

Ojeda-Magaña

Quintanilla-Domínguez

Ruelas

et al. 2014

Geoderma

Self Cite

View full text Add to dashboard Cite

Recent advances in non-destructive imaging techniques, such as X-ray computed tomography (CT), make it possible to analyse pore space features from the direct visualisation from soil structures. A quantitative characterisation of the three-dimensional solid-pore architecture is important to understand soil mechanics, as they relate to the control of biological, chemical, and physical processes across scales. This analysis technique therefore offers an opportunity to better interpret soil strata, as new and relevant information can be obtained. In this work, we propose an approach to automatically identify the pore structure of a set of 200-2D images that represent slices of an original 3D CT image of a soil sample, which can be accomplished through non-linear enhancement of the pixel grey levels and an image segmentation based on a PFCM (Possibilistic Fuzzy C-Means) algorithm. Once the solids and pore spaces have been identified, the set of 200-2D images is then used to reconstruct an approximation of the soil sample by projecting only the pore spaces. This reconstruction shows the structure of the soil and its pores, which become more bounded, less bounded, or unbounded with changes in depth. If the soil sample image quality is sufficiently favourable in terms of contrast, noise and sharpness, the pore identification is less complicated, and the PFCM clustering algorithm can be used without additional processing; otherwise, images require pre-processing before using this algorithm. Promising results were obtained with four soil samples, the first of which was used to show the algorithm validity and the additional three were used to demonstrate the robustness of our proposal. The methodology we present here can better detect the solid soil and pore spaces on CT images, enabling the generation of better 2D-3D representations of pore structures from segmented 2D images.

show abstract

“…Jovanovic et al (2013) proposed a segmentation scheme which can be performed already at the stage of sinograms. Cortina-Januchs et al (2011) used a segmentation/classification technique based on a combination of clustering and artificial neural network (ANN) to segment binary soil images, whereas Khan et al (2016) used the supervised technique least-squares support vector machine (LS-SVM) for segmentation of XCT rock images. Therefore, with the continuously, improving CT technologies and computational resources, machine learning (ML) techniques can be an effective tool for segment and classify for phase segmentation of XCT rock images.…”

Section: Introductionmentioning

confidence: 99%

Phase segmentation of X-ray computer tomography rock images using machine learning techniques: an accuracy and performance study

Chauhan

Rühaak

Anbergen³

et al. 2016

Solid Earth

View full text Add to dashboard Cite

Abstract. Performance and accuracy of machine learning techniques to segment rock grains, matrix and pore voxels from a 3-D volume of X-ray tomographic (XCT) grayscale rock images was evaluated. The segmentation and classification capability of unsupervised (k-means, fuzzy c-means, self-organized maps), supervised (artificial neural networks, least-squares support vector machines) and ensemble classifiers (bragging and boosting) were tested using XCT images of andesite volcanic rock, Berea sandstone, Rotliegend sandstone and a synthetic sample. The averaged porosity obtained for andesite (15.8 ± 2.5 %), Berea sandstone (16.3 ± 2.6 %), Rotliegend sandstone (13.4 ± 7.4 %) and the synthetic sample (48.3 ± 13.3 %) is in very good agreement with the respective laboratory measurement data and varies by a factor of 0.2. The k-means algorithm is the fastest of all machine learning algorithms, whereas a least-squares support vector machine is the most computationally expensive. Metrics entropy, purity, mean square root error, receiver operational characteristic curve and 10 K-fold cross-validation were used to determine the accuracy of unsupervised, supervised and ensemble classifier techniques. In general, the accuracy was found to be largely affected by the feature vector selection scheme. As it is always a trade-off between performance and accuracy, it is difficult to isolate one particular machine learning algorithm which is best suited for the complex phase segmentation problem. Therefore, our investigation provides parameters that can help in selecting the appropriate machine learning techniques for phase segmentation.

show abstract

Detection of pore space in CT soil images using artificial neural networks

Abstract: Abstract. Computed Tomography

Cited by 40 publications

References 22 publications

Pore detection in 3‐D CT soil samples through an improved sub‐segmentation method

Pore detection in 3‐D CT soil samples through an improved sub‐segmentation method

Identification of pore spaces in 3D CT soil images using PFCM partitional clustering

Phase segmentation of X-ray computer tomography rock images using machine learning techniques: an accuracy and performance study

Contact Info

Product

Resources

About