Artificial intelligence methods have a very wide range of applications. From speech recognition to self-driving cars, the development of modern deep-learning architectures is helping researchers to achieve new levels of accuracy in different fields. Although deep convolutional neural networks (CNNs) (a kind of deep-learning technique) have reached or surpassed human-level performance in image recognition tasks, little has been done to transport this new image classification technology to geoscientific problems. We have developed what we believe to be the first use of CNNs to identify lithofacies in cores. We use highly accurate models (trained with millions of images) and transfer learning to classify images of cored carbonate rocks. We found that different modern CNN architectures can achieve high levels of lithologic image classification accuracy (approximately 90%) and can aid in the core description task. This core image classification technique has the potential to greatly standardize and accelerate the description process. We also provide the community with a new set of labeled data that can be used for further geologic/data science studies.
We have determined how stratigraphy and lithofacies control pore structures in the Mississippian limestone and chert reservoir of north-central Oklahoma. There are 17 lithofacies and 29 high-frequency cycles documented in the Mississippian interval of this study. The high-frequency cycles have thicknesses ranging from 0.3 to 30.5 m (1–100 ft) and are mainly asymmetric regressive phases. The pore characteristics, measured through digital-image analysis (DIA) of thin-sections photomicrographs ([Formula: see text]100), exhibit unique correlations with core porosity, permeability, and lithofacies within a sequence-stratigraphic framework. There are five fundamental correlations observed. First, porosity from DIA and laboratory core measurements has a strong positive relationship ([Formula: see text]). However, some values from DIA porosity yield relatively higher values, specifically in spiculitic mudstone wackestones and argillaceous spiculitic mudstone wackestones. The difference is hypothesized due to the presence of isolated nanopores that are not accessible by helium during measurement of core porosity. Second, the relationship between pore circularity and permeability is indeterminate. The indeterminate relationship is related to a complex internal pore network, intensive diagenetic alteration, an unconnected microfracture network, and isolated pores. Third, positive moderate to strong correlations ([Formula: see text]) between porosity and permeability are observed only in four lithofacies. Fourth, coarse-grained lithofacies within the uppermost depositional sequence of the Mississippian interval have a heterogeneous pore-size distribution, whereas fine-grained lithofacies tend to exhibit a homogeneous pore-size distribution. Fifth, higher reservoir quality is associated with the upper intervals of high-frequency shallowing-upward cycles. This confirms that the sequence-stratigraphic variability of lithofacies is important to predict reservoir quality and its distribution. An alternative graphical method of pore-size distribution is also developed. To be a useful “technique,” examples of the plot are demonstrated using samples in this study. The plot successfully provides simple identification of pore-size classes, quantitative percentage of pore-size class, dominant pore class, and approximate minimum and maximum pore size.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.