Technology Focus: Digital Data Acquisition (January 2022)

Abstract

This third installment of the Digital Data Acquisition Technology Focus will focus on computer vision (CV) for improved data acquisition. CV is defined as a field of study that seeks to develop techniques to help computers “see” and understand the content of digital images such as photographs and videos. The field of computer vision has seen significant growth in recent years because of the convergence of multiple events, such as advancements in machine-learning technology (e.g., deep learning), an exponential increase in availability of visual data from mobile phones and platforms such as YouTube, and the growth of new use cases such as autonomous driving. As such, it is no wonder that CV is also having a major effect on the oil and gas industry, reflected in the number of papers this year on this subject alone. The oil and gas industry is awash with image and video data such as core images, seismic profiles, maps, video feeds for monitoring remote operations, and, these days, even drone images. The traditional approach to interpreting and acting on this data has been highly manual, which can be extremely time-consuming and prone to bias and error. CV technology intends to automate these processes and significantly improve the turnaround time from data to decisions. The papers chosen this year include application of CV for reservoir characterization, such as source-rock reconstruction, facies identification, and seismic data enhancement; and safety applications such as rig-state identification and automated corrosion mapping. Given the versatility of CV technology, its application across the oil and gas industry clearly will continue to grow. Recommended additional reading at OnePetro: www.onepetro.org. SPE 204216 - Deep-Learning-Based Vuggy Facies Identification From Borehole Images by Jiajun Jiang, Baylor University, et al. SPE 202710 - Machine-Learning-Based Seismic Data Enhancement Toward Overcoming Geophysical Limitations by Shotaro Nakayama, INPEX, et al. SPE 205347 - Machine-Learning-Assisted Segmentation of Focused Ion Beam Scanning Electron Microscopy Images With Artifacts for Improved Void-Space Characterization of Tight Reservoir Rocks by Andrey Kazak, Skolkovo Institute of Science and Technology, et al.