Two intelligent pattern recognition models for automatic identification of textural and pore space characteristics of the carbonate reservoir rocks using thin section images

Borazjani, Omid; Ghiasi-Freez, Javad; Hatampour, Amir

doi:10.1016/j.jngse.2016.09.048

Cited by 16 publications

(35 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Feature selection appears to be related more to the ease of acquisition rather than any proven utility. Most studies do not undertake visualization of the data in hyperspace using PCA (Abedini et al, 2018;Borazjani et al, 2016;Ghiasi-Freez et al, 2012;Mollajan et al, 2016;Z. Wang et al, 2022), thereby obfuscating the underpinning drivers of their reported excellent model accuracies.…”

Section: Study Design Issues In Related Studiesmentioning

confidence: 99%

An Object‐Based Approach to Differentiate Pores and Microfractures in Petrographic Analysis Using Explainable, Supervised Machine Learning

Jayachandran,

Gibbs,

Laya

et al. 2024

Earth and Space Science

View full text Add to dashboard Cite

Petrographic observations are vital for carbonate pore‐typing, linking geological frameworks to petrophysical behavior. However, current petrographic pore typing is manual, with the qualitative to semi‐quantitative results not easily fitted into quantitative subsurface characterization. Some recent studies have automated this process using supervised machine learning (ML) and deep learning (DL), focusing on simple pore morphological features, and have reported high classification accuracies for several complex pore types. However, there are concerns about the validity of these studies due to conceptual and technical flaws in their collective approach. This study was aimed at a more fundamental problem, classifying between open microfractures and open pores in petrographic thin sections using an object‐based approach and explainable supervised ML. We analyzed 18 carbonate thin sections from the USA, numerically representing them using five shape features: compactness, aspect ratio, extent, solidity, and formfactor. Using a labeled data set of 400 microfractures and 400 pores, we evaluated nine of the most widely used supervised models. All models showed high testing accuracies (89.58%–90.42%). Interestingly, complex non‐linear models did not significantly outperform simpler linear ones. Compactness and aspect ratio were the most informative features. However, the labeled data sets did not reflect the overall data set's complexity, which suggested that high accuracies in similar studies might be due to curated data sets rather than accounting for the true complexity of carbonate pore systems. The study concludes that simple shape features are ineffective for classifying carbonate pore types. It is hoped that this study will provide a foundation for more robust artificial intelligence‐assisted pore typing.

show abstract

Section: Study Design Issues In Related Studiesmentioning

confidence: 99%

An Object‐Based Approach to Differentiate Pores and Microfractures in Petrographic Analysis Using Explainable, Supervised Machine Learning

Jayachandran,

Gibbs,

Laya

et al. 2024

Earth and Space Science

View full text Add to dashboard Cite

show abstract

“…These parameters identify and measure the pore spot in cast thin-section images, which are thin-sections impregnated with colored epoxy. Based on their unique color, pores can be identified by threshold methods in the RGB or HSV color spaces [5,6]. In addition, pattern recognition and GIS-based methods are applied to extract the boundary and region of the pore as a polygon object, and, further, to quantitatively calculate its shape, orientation, type, and spatial distribution [7][8][9][10].…”

Section: A Pore Information Extractionmentioning

confidence: 99%

Rock classification in petrographic thin section images based on concatenated convolutional neural networks

Zhu

et al. 2020

Earth Sci Inform

View full text Add to dashboard Cite

Rock classification plays an important role in rock mechanics, petrology, mining engineering, magmatic processes, and numerous other fields pertaining to geosciences. This study proposes a concatenated convolutional neural network (Con-CNN) method for classifying the geologic rock type based on petrographic thin sections. Herein, plane polarized light (PPL) and crossed polarized light (XPL) were used to acquire thin section images as the fundamental data. After conducting the necessary pre-processing analyses, the PPL and XPL images as well as their comprehensive image (CI) were incorporated in three convolutional neural networks (CNNs) comprising the same structure for achieving a preliminary classification; these images were developed by employing the fused principal component analysis (PCA). Subsequently, the results of the CNNs were concatenated by using the maximum likelihood detection to obtain a comprehensive classification result. Finally, a statistical revision was applied to fix the misclassification due to the proportion difference of minerals that were similar in appearance. In this study, 13 types of 92 rock samples, 196 petrographic thin sections, 588 images, and 63504 image patches were fabricated for the training and validation of the Con-CNN. The five-folds cross validation shows that the method proposed provides an overall accuracy of 89.97%, which facilitates the automation of rock classification in petrographic thin sections.

show abstract

“…X-ray CT can achieve wider field of view and very high resolution in 3D space (Ge et al, 2015). Despite the lower resolution in comparison with µxCT and SEM images, the thin section images have other advantages such as a wider range of view, saved time and cheaper availability for petrophysical studies, such as pore microstructure (Desbois et al, 2011;Rabbani et al, 2014a;Borazjani et al, 2016;Gundogar et al, 2016;Rabbani et al, 2016;Xiao et al, 2016), mineral recognition and classification (Hofmann et al, 2013;Asmussen et al, 2015;Izadi et al, 2015;Izadi et al, 2017b), specific surface area (Rabbani and Jamshidi, 2014;Rabbani et al, 2014b), elastic modulus (Arns et al, 2002;Dvorkin et al, 2011;Madonna et al, 2012;Saxena and Mavko, 2016), rock type determination (Mynarczuk, 2010;Mynarczuk et al, 2013;Ge et al, 2015;Mollajan et al, 2016), pore-grain analysis (Rabbani and Jamshidi, 2014;Rabbani et al, 2014b;Song et al, 2016), flowing property (Peng et al, 2016;Wang et al, 2016b). Using thin section images, the interconnected pore structure can be marked out visually as they are filled by color epoxy resin.…”

Section: Petrophysical Characterization Based On Thin Section Analysismentioning

confidence: 99%

Applications of digital core analysis and hydraulic flow units in petrophysical characterization

Chen

Zhou

2017

Adv. Geo-Energ. Res.

View full text Add to dashboard Cite

Conventional petrophysical characterizations are often based on direct laboratory measurements. Although they provide accurate results, such measurements are time-consuming and limited by instrument and environment. What's more, in the georesource energy industry, availability and cuttings of core plugs are difficult. Because of these reasons, virtual digital core technology is of increasing interest due to its capability of characterizing rock samples without physical cores and experiments. Virtual digital core technology, also known as digital rock physics, is used to discover, understand and model relationships between material, fluid composition, rock microstructure and macro equivalent physical properties. Based on actual geological conditions, modern mathematical methods and imaging technology, the digital model of the core or porous media is created to carry out physical field numerical simulation. In this review, the methods for constructing digital porous media are introduced first, then the characterization of thin rock cross section and the capillary pressure curve using scanning electron microscope image under mixed wetting are presented. Finally, we summarize the hydraulic flow unit methods in petrophysical classification.Keywords: Digital core, porous media, petrophysical characterization, thin section, mercury injection capillary pressure.Citation: Chen, X., Zhou, Y. Applications of digital core analysis and hydraulic flow units in petrophysical characterization. Adv.

show abstract

Two intelligent pattern recognition models for automatic identification of textural and pore space characteristics of the carbonate reservoir rocks using thin section images

Cited by 16 publications

References 10 publications

An Object‐Based Approach to Differentiate Pores and Microfractures in Petrographic Analysis Using Explainable, Supervised Machine Learning

An Object‐Based Approach to Differentiate Pores and Microfractures in Petrographic Analysis Using Explainable, Supervised Machine Learning

Rock classification in petrographic thin section images based on concatenated convolutional neural networks

Applications of digital core analysis and hydraulic flow units in petrophysical characterization

Contact Info

Product

Resources

About