Multimodal Biometrics using adaptive combination of many features

2019 Fiscal Year Final Research Report

• Project/Area Number: 16K18103
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Communication/Network engineering
• Research Institution: The University of Electro-Communications
• Principal Investigator: Ichino Masatsugu 電気通信大学, 大学院情報理工学研究科, 准教授 (80548892)
• Project Period (FY): 2016-04-01 – 2020-03-31
• Keywords: 生体認証

Outline of Final Research Achievements

Biometrics recognizes the system user and becomes more important. It makes advances in technology of camera and enables to shoot a video sequence of high resolution. And it enables to utilize dozens of modalities at the same time. Therefore, I focused on the authentication method to integrate many features of each modality. In this research, I positively utilized the individuality of each area around the eyes. I proposed iris and periocular authentication using divided images around the eyes. As a result, compared with previous research, the authentication accuracy was improved, and the effectiveness of the proposed method using region partitioning could be confirmed.

Free Research Field

セキュリティ

Academic Significance and Societal Importance of the Research Achievements

システム利用者の正当性を確認する生体認証の重要性が高まる中，複数のモダリティを用いるマルチモーダルバイオメトリクスが実用化に向けての現実的な解として認識されている．そこで本研究では，マルチモーダルバイオメトリクスで必要となる特徴量を組み合わせる方法に着目した．本研究での提案手法は，モダリティの個人性を考慮し，多くの特徴量を組み合わせた認証を行うことを可能とした．この成果をさらに発展させることで，生体認証におけるマルチモーダルバイオメトリクスの適用範囲を広げることができると考える．