Super-resolution imaging for understanding hierarchical structures produced by water-rock reactions and mass transfer

2019 Fiscal Year Final Research Report

• Project/Area Number: 18K18778
• Research Category: Grant-in-Aid for Challenging Research (Exploratory)
• Allocation Type: Multi-year Fund
• Review Section: Medium-sized Section 17:Earth and planetary science and related fields
• Research Institution: Tohoku University
• Principal Investigator: Okamoto Atsushi 東北大学, 環境科学研究科, 准教授 (40422092)
• Co-Investigator(Kenkyū-buntansha): 大森 敏明 神戸大学, 工学研究科, 准教授 (10391898)
• Project Period (FY): 2018-06-29 – 2020-03-31
• Keywords: X線CT / スパース超解像 / 掘削コア / 岩石組織 / 蛇紋岩化作用 / オマーン掘削プロジェクト / 岩石ー水相互作用

Outline of Final Research Achievements

bbbX-ray tomography (CT) is a non-destructive method for investigating three-dimensional structures. In recent years, this method has been widely applied to rock drill core samples. However, when imaging a wide area, there is a problem that the resolution is not enough for analyzing microstructure of rocks. In this research, I challenged to develop a new method of decoding rock textures that connects micro and macro, based on "sparse super-resolution technology". The analyzed sample was a serpentinized peridotite taken from the Oman Drilling Project, and the same part of the rock core was imaged at different resolutions. By sparse super-resolution, a small number of characteristic (sparse) basis images were extracted, and high-performance super-resolution was successful compared to conventional interpolation methods. Furthermore, it was found that stepwise super-resolution is effective when there is a large difference in resolution.

Free Research Field

岩石学

Academic Significance and Societal Importance of the Research Achievements

地球内部では、鉱物からプレートまで様々なスケールの現象が連動して起こっている。近年、分析・解析装置は飛躍的に向上たものの、異なるスケールの間を結びつけることは未だにできていない。本研究では、X線CTという岩石・地質試料の普遍的に使われる分析方法に、スパース超解像という機械学習技術の１つを適用して岩石組織解読の方法論を提案したものであり、この「スケールの壁」を打破すること第一歩になったと考えられる。