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

Signal Recognition Mechanisms by Selecting Higher-Order Spectral Features Through Learning

Research Project

Project/Area Number 16K00322
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Research Field Soft computing
Research InstitutionUniversity of Tsukuba

Principal Investigator

Kameyama Keisuke  筑波大学, システム情報系, 教授 (40242309)

Project Period (FY) 2016-04-01 – 2020-03-31
Project Status Completed (Fiscal Year 2019)
Budget Amount *help
¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000)
Fiscal Year 2018: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2017: ¥2,080,000 (Direct Cost: ¥1,600,000、Indirect Cost: ¥480,000)
Fiscal Year 2016: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Keywords高次スペクトル / 信号認識 / ニューラルネットワーク / 学習 / 虹彩認証 / 超解像 / サポートベクトルマシン / カーネル法 / アンサンブル学習 / 転移学習 / 生体認証 / 虹彩 / パターン認識 / 特徴抽出 / 高次自己相関
Outline of Final Research Achievements

Higher-Order Spectral features of signals provide nonlinear signal features unavailable in power spectra. However, their high-dimensionality, sparseness and high computational cost were obstacles for general use for signal classification. In this work, we attempted to solve this issue by joint use of Local Higher-Order Moment Spectrum (LHOMS) kernel functions and gradient based learning of neural networks. A modified image feature extraction method by LHOMS kernel was applied to iris authentication. It was found that the method allows authentication robust to additive noise to the iris image. Also, a novel method for inheriting the probability density of parameters in transfer learning of convolutional neural networks was introduced. The novel transfer learning improved the training efficiency when compared with the existing methods.

Academic Significance and Societal Importance of the Research Achievements

本研究で得られた成果は,画像や音などのメディア信号の認識に際して用いる特徴量として,高次スペクトル特徴量の抽出とニューラルネットワークを用いた分類学習に関する新たな知見を提供するものである.提案手法は生体認証や画像認識の問題に適用され,それぞれその有効性を示している.メディア信号の学習に基づく自動認識や分類はその重要性がますます増大しており,本研究の成果はそれらを行う手段としてに新たな選択肢を提供するものである.

Report

(5 results)
  • 2019 Annual Research Report   Final Research Report ( PDF )
  • 2018 Research-status Report
  • 2017 Research-status Report
  • 2016 Research-status Report
  • Research Products

    (11 results)

All 2019 2018 2017 2016

All Journal Article (3 results) (of which Int'l Joint Research: 1 results,  Peer Reviewed: 3 results) Presentation (8 results) (of which Int'l Joint Research: 1 results)

  • [Journal Article] Parallel Cooperative Ensemble Learning by Adaptive Data Weighting and Error-Correcting Output Codes2018

    • Author(s)
      Shota Utsumi and Keisuke Kameyama
    • Journal Title

      Lecture Notes in Computer Science

      Volume: 11303 Pages: 673-683

    • DOI

      10.1007/978-3-030-04182-3_59

    • ISBN
      9783030041816, 9783030041823
    • Related Report
      2018 Research-status Report
    • Peer Reviewed
  • [Journal Article] Parameter Density Inheritance Using Kernel Density Estimation for Efficient CNN Learning2018

    • Author(s)
      Keisuke Horiuchi and Keisuke Kameyama
    • Journal Title

      Proc. 2018 IEEE International Symposium on Signal Processing and Information Technology

      Volume: 1 Pages: 308-313

    • DOI

      10.1109/isspit.2018.8642618

    • Related Report
      2018 Research-status Report
    • Peer Reviewed
  • [Journal Article] Automatic MOOC video classification using transcript features and convolutional neural networks2017

    • Author(s)
      Houssem Chatbri, Kevin McGuinness, Suzanne Little, Jiang Zhou, Keisuke Kameyama, Paul Kwan and Noel O'Connor
    • Journal Title

      ACM Multimedia 2017 - MultiEdTech Workshop

      Volume: 1 Pages: 21-26

    • DOI

      10.1145/3132390.3132393

    • Related Report
      2017 Research-status Report
    • Peer Reviewed / Int'l Joint Research
  • [Presentation] CNN を用いた低解像度虹彩画像からの認証特徴量予測2019

    • Author(s)
      渡邉 崚,亀山啓輔
    • Organizer
      電子情報通信学会バイオメトリクス研究会
    • Related Report
      2019 Annual Research Report
  • [Presentation] 並列的なアンサンブル学習におけるデータの重みを用いた相補的な弱学習器の構成2018

    • Author(s)
      内海翔太,亀山啓輔
    • Organizer
      電子情報通信学会IBISML研究会
    • Related Report
      2018 Research-status Report
  • [Presentation] カーネル密度関数推定を用いたパラメータ密度の継承によるCNNの学習効率化2018

    • Author(s)
      堀内圭佑, 亀山啓輔
    • Organizer
      電子情報通信学会IBISML研究会
    • Related Report
      2018 Research-status Report
  • [Presentation] アンサンブル学習における弱学習器間の相互性指標に基づく学習法の修正に関する一検討2018

    • Author(s)
      内海翔太, 亀山啓輔
    • Organizer
      電子情報通信学会IBISML研究会
    • Related Report
      2018 Research-status Report
  • [Presentation] Iris Authentication using Local Spectral Features and their Relational Operations2017

    • Author(s)
      Miharu Aizawa, Keisuke Kameyama
    • Organizer
      International Workshop on Advanced Image Technology
    • Place of Presentation
      マレーシア ペナン島
    • Year and Date
      2017-01-08
    • Related Report
      2016 Research-status Report
    • Int'l Joint Research
  • [Presentation] 局所高次スペクトル特徴量の関係演算による虹彩コード生成2016

    • Author(s)
      相澤美晴,亀山啓輔
    • Organizer
      バイオメトリクスと認識・認証シンポジウム
    • Place of Presentation
      芝浦工業大学豊洲キャンパス(東京都港区)
    • Year and Date
      2016-11-16
    • Related Report
      2016 Research-status Report
  • [Presentation] 局所スペクトル特徴量とその関係演算を用いた虹彩認証2016

    • Author(s)
      相澤美晴,亀山啓輔
    • Organizer
      電子情報通信学会バイオメトリクス研究会
    • Place of Presentation
      大阪電気通信大学(大阪府寝屋川市)
    • Related Report
      2016 Research-status Report
  • [Presentation] 層状ニューラルネットワーク2クラス分類器の組み合わせによる多クラス分類器に関する考察2016

    • Author(s)
      内海翔太,亀山啓輔
    • Organizer
      電子情報通信学会ニューロコンピューティング研究会
    • Place of Presentation
      大阪電気通信大学(大阪府寝屋川市)
    • Related Report
      2016 Research-status Report

URL: 

Published: 2016-04-21   Modified: 2021-02-19  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi