• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to project page

2023 Fiscal Year Final Research Report

From artisanship to amateur: the challenge to identify large fossils using deep learning

Research Project

  • PDF
Project/Area Number 21K14031
Research Category

Grant-in-Aid for Early-Career Scientists

Allocation TypeMulti-year Fund
Review Section Basic Section 17050:Biogeosciences-related
Research InstitutionKyushu University

Principal Investigator

Matsui Kumiko  九州大学, 総合研究博物館, 特別研究員(CPD) (80816207)

Project Period (FY) 2021-04-01 – 2024-03-31
Keywords化石 / 種同定 / 軟体動物
Outline of Final Research Achievements

This research project attempts to automate the identification of fossil species. Species identification of fossils is fundamental in paleontology and is an essential technique for determining geological ages and estimating paleo and sedimentary environments. Fossil species are identified based on morphological descriptions, but in many cases, these identifications have been made subjectively based on qualitative indicators. Accurate identification requires years of experience and has reduced the efficiency of paleontological research. Therefore, this study first digitizes fossils into 3D digital data. Then, by processing the obtained data using deep learning technologies, the project aims to automate fossil species classification, significantly simplifying the identification process in paleontology.

Free Research Field

古生物学

Academic Significance and Societal Importance of the Research Achievements

現生生物においてはDNA情報を用いる種同定自動化技術は現在盛んに研究されているものの,形態データにディープラーニングといった人工知能技術を用いた例は非常に限られており,かつ大型化石を種レベルで同定しようという試みは少ない.本研究で使用したイノセラムス類化石は複雑な立体形状を持ち、かつ変形しているなど,形態情報が多い分類群であり,分類のために利用できる情報が限定的で深層学習技術を導入するには比較的悪い条件の対象群であった.これまで「職人技」であった形態種の同定技術をこれまでよりも確実に多くの人が同一の基準を持って実施できるような技術へと変換する技術を提供するものである.

URL: 

Published: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi