GIS and Machine Learning: A New Approach to Discovering Archaeological Sites

2022 Fiscal Year Final Research Report

• Project/Area Number: 21K18408
• Research Category: Grant-in-Aid for Challenging Research (Exploratory)
• Allocation Type: Multi-year Fund
• Review Section: Medium-sized Section 4:Geography, cultural anthropology, folklore, and related fields
• Research Institution: Nara National Research Institute for Cultural Properties
• Principal Investigator: Takata Yuichi 独立行政法人国立文化財機構奈良文化財研究所, 企画調整部, 主任研究員 (50708576)
• Co-Investigator(Kenkyū-buntansha): 野口 淳 金沢大学, 古代文明・文化資源学研究所, 客員研究員 (70308063)
• Project Period (FY): 2021-07-09 – 2023-03-31
• Keywords: 機械学習 / DEM / GIS / 遺跡 / 画像解析 / 考古学ビッグデータ / 古墳

Outline of Final Research Achievements

In this research, by combining high-precision topographical digital data, a machine learning image analysis program, and a huge amount of known ruins information, we automatically extract candidates for newly discovered ruins on GIS and conduct field surveys based on them. By doing so, we will develop a method for discovering new ruins. In order to achieve this, we set three objectives: the preparation of archaeological site information (position and area) (Purpose I), the creation and processing of archaeological site image analysis programs (Purpose II), and the discovery of new archaeological sites and the development of methods (Purpose III). As a result of the research, we created a newly discovered archaeological site model and extracted candidate archaeological sites from high-precision topographical data. During the actual field survey, we discovered several undiscovered burial mounds.

Free Research Field

考古学

Academic Significance and Societal Importance of the Research Achievements

遺跡の踏査は文化財保護の根幹でもあるが負担が大きいため、実施が難しい場合がある。そのため山間部の遺跡が未発見の場合がある。開発事業を前提とした踏査は、遺跡が新発見されても破壊前提となる場合が多い。新たな遺跡の発見は、地域の歴史を詳らかにするうえで重要であり意義が大きい。本研究で効率的に遺跡を発見する手法を確立できれば、歴史研究を加速させ、文化財保護にも貢献できる。高精度な地理データ、画像解析プログラム、既知の大量の遺跡情報の組み合わせは、既存の遺跡踏査手法を飛躍的に向上させ、調査手法のブレイクスルーとなる可能性がある。