Abstract
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.
Highlights
Micro X-ray computer tomography (XCT) images of a rock sample help in classification of pore space and assist in modeling of pore-network geometries
Qi are the images used for training the forward artificial neural network (FFANN) (k-means and fuzzy c-means (FCMs) images), Qi is the mean of the images used for training FFANN and Qi is the mean of the classified images To evaluate accuracy of our FFANN model, we looked at the mean square relative error (MSRE) values
The porosities which were determined from the stack of 10 XCT slices for three to seven classes using different machine learning (ML) techniques are shown in the Fig. 2
Summary
Micro X-ray computer tomography (XCT) images of a rock sample help in classification of pore space and assist in modeling of pore-network geometries. Porous materials like sedimentary and volcanic rocks contain areas of void, called the pore space, as well as a number of distinct mineral components, each with a comparatively uniform density. These different components are referred to as phases. Segmentation of a porous rock means deciding to which phase each voxel belongs Tomographic images of such materials consist of a cubic array of reconstructed linear X-ray attenuation coefficient values each corresponding to a voxel of the sample. The focus of this study is to assess the performance and accuracy of the above mentioned ML techniques to segment rock grain, matrix and pore phases in heterogeneous rock samples such as andesite, Berea sandstone, Rotliegend sandstone and a synthetic sample containing microporosities
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
