An Universal Automatic Signature Verification with Explainability

Research Project

Project/Area Number	21K11942
Research Category	Grant-in-Aid for Scientific Research (C)
Allocation Type	Multi-year Fund
Section	一般
Review Section	Basic Section 61010:Perceptual information processing-related
Research Institution	Tokyo Denki University (2022-2023) Saitama Institute of Technology (2021)
Principal Investigator	Ohyama Wataru 東京電機大学, システムデザイン工学部, 教授 (10324550)
Project Period (FY)	2021-04-01 – 2024-03-31
Project Status	Completed (Fiscal Year 2023)
Budget Amount *help	¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000) Fiscal Year 2023: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000) Fiscal Year 2022: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000) Fiscal Year 2021: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
Keywords	バイオメトリクス / 署名照合 / 機械学習 / バイオメトリクス認証 / 生体認証 / パターン認識
Outline of Research at the Start	機械学習技術を活用した署名照合の自動化には，（１）照合判定根拠に対する説明性の向上，（２）照合精度の向上，（３）様々な言語の署名を統一的に照合できるユニバーサル性の向上といった課題が残されている．本研究は，主に以下の３項目について研究を行い，有用かつ実用的な自動署名照合の確立を目指す．（１）機械学習を活用した自動署名照合における照合判定根拠を説明できる技術の開発，（２）応募者らが開発した組み合わせ分割署名照合や，応募者らが初めて署名照合に適用したランキング学習等を活用した自動署名照合の高度化，（３）様々な言語の署名を統一的に照合する「ユニバーサル署名照合」の実現
Outline of Final Research Achievements	The automatic signature verification using machine learning technology still faces challenges in (1) improving the explainability of the verification decision basis, (2) enhancing verification accuracy, and (3) increasing the universality to uniformly verify signatures in various languages. This study primarily addresses (1) the development of technology that can explain the basis of verification decisions in automated signature verification using machine learning, (2) the advancement of automated signature verification by utilizing methods such as combinatorial partitioned signature verification, ranking learning, and handwriting verification focusing on local variations, and (3) the realization of a "universal signature verification" that can uniformly verify signatures in various languages.
Academic Significance and Societal Importance of the Research Achievements	署名照合は国際的には社会的に広く受け入れられている本人確認手法である．本研究の成果は，署名照合の自動化に残されていた上述の課題を解決する糸口になることが期待される．