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

2022 Fiscal Year Final Research Report

Image feature extraction based on wavelet, topological data analysis, deep learning, and its theory

Research Project

  • PDF
Project/Area Number 19K03623
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Review Section Basic Section 12040:Applied mathematics and statistics-related
Research InstitutionSaga University

Principal Investigator

Minamoto Teruya  佐賀大学, 理工学部, 教授 (00294900)

Project Period (FY) 2019-04-01 – 2023-03-31
Keywordsウェーブレット解析 / 位相的データ解析 / 早期がん検出 / 電子透かし / 歩容
Outline of Final Research Achievements

Focusing on the multi-scale property of wavelet and topological data analysis, we developed methods to extract interpretable "high-quality features" from images. Then, by combining these features with machine learning methods, we developed various image processing and analysis methods. Specifically, we developed methods for detecting early-stage colorectal cancer and a non-reference watermarking method based on the dual-tree complex discrete Wavelet transform, methods for detecting gait and stained asbestos based on the Dyadic Wavelet transform, a method for detecting dysplasia using Lifting Wavelets, and a method for detecting early gastric cancer using Curvelets.

Free Research Field

応用数学

Academic Significance and Societal Importance of the Research Achievements

本研究により、ウェーブレット解析と位相的データ解析のマルチスケール性を活用して画像から人間が解釈可能な特徴量を抽出する手法を開発し、機械学習モデルの説明可能性の向上にも寄与した。特に、早期がん検出法の開発は医療分野での早期診断に貢献し、患者の治療成功率と生存率の向上につながる。また、非参照型電子透かし法や歩行検出などの開発は、セキュリティや監視システムの向上に寄与する。これらの成果は科学的知見の進歩だけでなく、医療、セキュリティ、社会全体の福祉向上にも寄与すると言える。

URL: 

Published: 2024-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi