Machine-Learning-Assisted Segmentation of Focused Ion Beam-Scanning Electron Microscopy Images with Artifacts for Improved Void-Space Characterization of Tight Reservoir Rocks

Abstract

Summary The modern focused ion beam-scanning electron microscopy (FIB-SEM) allows imaging of nanoporous tight reservoir-rock samples in 3D at a resolution up to 3 nm/voxel. Correct porosity determination from FIB-SEM images requires fast and robust segmentation. However, the quality and efficient segmentation of FIB-SEM images is still a complicated and challenging task. Typically, a trained operator spends days or weeks in subjective and semimanual labeling of a single FIB-SEM data set. The presence of FIB-SEM artifacts, such as porebacks, requires developing a new methodology for efficient image segmentation. We have developed a method for simplification of multimodal segmentation of FIB-SEM data sets using machine-learning (ML)-based techniques. We study a collection of rock samples formed according to the petrophysical interpretation of well logs from a complex tight gas reservoir rock of the Berezov Formation (West Siberia, Russia). The core samples were passed through a multiscale imaging workflow for pore-space-structure upscaling from nanometer to log scale. FIB-SEM imaging resolved the finest scale using a dual-beam analytical system. Image segmentation used an architecture derived from a convolutional neural network (CNN) in the DeepUNet (Ronneberger et al. 2015) configuration. We implemented the solution in the Pytorch® (Facebook, Inc., Menlo Park, California, USA) framework in a Linux environment. Computation exploited a high-performance computing system. The acquired data included three 3D FIB-SEM data sets with a physical size of approximately 20 × 15 × 25 µm with a voxel size of 5 nm. A professional geologist manually segmented (labeled) a fraction of slices. We split the labeled slices into training, validation, and test data. We then augmented the training data to increase its size. The developed CNN delivered promising results. The model performed automatic segmentation with the following average quality indicators according to test data: accuracy of 86.66%, precision of 54.93%, recall of 83.76%, and F1 score of 55.10%. We achieved a significant boost in segmentation speed of 14.5 megapixel (MP)/min. Compared with 0.18 to 1.45 MP/min for manual labeling, this yielded an efficiency increase of at least 10 times. The presented research work improves the quality of quantitative petrophysical characterization of complex reservoir rocks using digital rock imaging. The development allows the multiphase segmentation of 3D FIB-SEM data complicated with artifacts. It delivers correct and precise pore-space segmentation, resulting in little turn-around-time saving and increased porosity-data quality. Although image segmentation using CNNs is mainstream in the modern ML world, it is an emerging novel approach for reservoir-characterizationtasks.

Review of the Development Status and Technology of Tight Oil: Advances and Outlook

“…According to BP’s Energy Outlook 2030 report, North America has approximately 10 billion tons of tight oil resources, making it the region with the richest proven tight oil resources in the world . Tight oil resources are mainly distributed along the Rocky Mountains and Appalachian Mountains. − The United States is currently the most successful country in extracting tight oil . Since the comprehensive application of horizontal well staged fracturing technology in 2006, the development of tight reservoirs has promoted the recovery of the North American oil supply. − At present, the main tight oil producing areas in the United States include the Bakun Formation, Naiebral, Barnett, and Eagle Beach tight oil areas.…”

A novel machine learning model for autonomous analysis and diagnosis of well integrity failures in artificial-lift production systems

“…The evaluation of model performance is related to its ability to predict on hold-out test data (Hastie et al, 2017). A confusion matrix is simply a combination of the actual status and model predictions used to measure model performance (Kazak et al, 2021). It is a special table structure that enables for the evaluation of an algorithm's performance.…”

Interactive Machine Learning Improves Accuracy of Coal Porosity Segmentation in Focused Ion Beam–Scanning Electron Microscopy Images

“…The most common way to segment the images produced during FIB–SEM analyses is to use grayscale thresholding, which filters pixels based on their intensity: dark pixels are assumed to represent pores, while bright pixels represent solid phases. ,,, However, the inherent complexity of the coal microstructure, coupled with image noise and artifacts means that grayscale segmentation can produce poor results. ,− Three particular effects are responsible for most of the errors during segmentation: (1) the pore-back effect, which occurs when a material inside a porenot within the current plane of interestappears in the image; the effect results in the pore region having the same grayscale value as the solid phase; , (2) charging effects, which arise as a result of charge accumulation on the specimen surface; the effect is common when imaging non-conductive materials, , and can cause a significantly higher grayscale value in the pore area, giving it a similar appearance to that of a “mineral” phase; (3) curtaining effects, which occur due to ion-beam milling; , the effect generates visible stripes with grayscale values comparable to those of the pore phase, leading to incorrect segmentation as pores, particularly when a higher grayscale thresholding level is used.…”