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

2018 Fiscal Year Annual Research Report

Development of Cellular Neural Networks with Complex Network Structures and their Applications to Information Processing

Research Project

Project/Area Number 16K06357
Research InstitutionThe University of Tokushima

Principal Investigator

西尾 芳文  徳島大学, 大学院社会産業理工学研究部(理工学域), 教授 (80253227)

Co-Investigator(Kenkyū-buntansha) 上手 洋子  徳島大学, 大学院社会産業理工学研究部(理工学域), 准教授 (80582642)
細川 康輝  四国大学, 経営情報学部, 准教授 (20341266)
Project Period (FY) 2016-04-01 – 2019-03-31
Keywordsセルラーニューラルネットワーク / 複雑ネットワーク / 非線形回路 / カオス
Outline of Annual Research Achievements

本研究では、研究代表者のグループがこれまでに行ってきた複雑系ネットワークの研究成果をセルラニューラルネットワーク(CNN)に組み込むことで、人間の脳での処理に近い高機能な情報処理をリアルタイムで行うことができる新しい複雑ネットワークCNNを開発する。
CNNがもつリアルタイム性とアナログ非線形性の利点を最大限に活かしつつ、複雑系ネットワークの特徴である創発性や時空間変動性を積極的に利用し、また、計算機シミュレーション・回路実験・理論解析により、システム設計・特性調査・アプリケーション開発・回路実装までを行うことで、様々な実問題に適用可能な高度な情報処理システムを実現することを目的としている。
平成30年度に得られた主たる成果は以下の通りである。
1.遅延層を含む二層CNNによる画像処理:本研究では、二層CNNの第二層に、遅延テンプレートを含むCNNをもつCNNの開発を行った。計算機シミュレーションにより、本提案手法が、複雑な入力画像のエッジ検出に効果があることを確認することができた。
2.二種類のテンプレートを含むCNN回路によるカラー画像処理:本研究では、二種類のテンプレートを市松模様に配置した二次元CNNを設計し、カラー画像の処理を行った。二種類のテンプレートをさまざまに変化させることで、入力画像のカラー情報に依存した処理を行うCNNを設計することができた。
3.複雑系カオスネットワークの同期:本研究では、カオス回路をノードに持つ複雑系ネットワークにおいて、ノード間の結合の強さがノードの次数によって変化する場合にみられる複雑な同期現象についての調査を行った。計算機シミュレーションにより、ノードの次数がネットワーク全体の同期状態に強く影響を与えることが明らかとなった。

  • Research Products

    (22 results)

All 2018 Other

All Int'l Joint Research (1 results) Journal Article (8 results) (of which Int'l Joint Research: 1 results,  Peer Reviewed: 8 results,  Open Access: 2 results) Presentation (13 results) (of which Int'l Joint Research: 13 results)

  • [Int'l Joint Research] Zurich University of Applied Sciences(スイス)

    • Country Name
      SWITZERLAND
    • Counterpart Institution
      Zurich University of Applied Sciences
  • [Journal Article] Synchronization Phenomena of Chaotic Circuit Networks with Distributed Hub Including Positive and Negative Coupling2018

    • Author(s)
      Shuhei HASHIMOTO, Yoko UWATE and Yoshifumi NISHIO
    • Journal Title

      Proceedings of IEEE International Symposium on Circuits and Systems

      Volume: - Pages: 1-5

    • DOI

      10.1109/ISCAS.2018.8351723

    • Peer Reviewed
  • [Journal Article] Producing Complex Networks Using Coupled Oscillatory Circuits with Evolutionary Connections2018

    • Author(s)
      Yoko UWATE, Thomas OTT and Yoshifumi NISHIO
    • Journal Title

      Proceedings of IEEE International Symposium on Circuits and Systems

      Volume: - Pages: 1-5

    • DOI

      10.1109/ISCAS.2018.8351665

    • Peer Reviewed / Int'l Joint Research
  • [Journal Article] Two-Layer Cellular Neural Networks with Layer of Delay Output2018

    • Author(s)
      Takahisa ANDO, Yoko UWATE and Yoshifumi NISHIO
    • Journal Title

      Proceedings of International Workshop on Cellular Nanoscale Networks and their Applications

      Volume: - Pages: 71-74

    • Peer Reviewed
  • [Journal Article] Learning Method Using Acetylcholine among Neurotransmitter in Feed Forward Neural Networks2018

    • Author(s)
      Kazuki NAGAO, Yoko UWATE and Yoshifumi NISHIO
    • Journal Title

      Proceedings of International Symposium on Nonlinear Theory and its Applications

      Volume: - Pages: 284-287

    • Peer Reviewed / Open Access
  • [Journal Article] Influence of Coupling Strengths on Synchronization Phenomena in Two Rings of Coupled van der Pol Oscillators2018

    • Author(s)
      Daiki NARIAI, Yoko UWATE and Yoshifumi NISHIO
    • Journal Title

      Proceedings of International Symposium on Nonlinear Theory and its Applications

      Volume: - Pages: 483-486

    • Peer Reviewed / Open Access
  • [Journal Article] Design of Two Template Cellular Neural Networks for Color Image Processing2018

    • Author(s)
      Yasuteru HOSOKAWA, Yoko UWATE and Yoshifumi NISHIO
    • Journal Title

      Proceedings of International SoC Design Conference

      Volume: - Pages: 162-163

    • Peer Reviewed
  • [Journal Article] Synchronization Phenomena of Coupled Chaotic Circuits Network with Coupling Strength Depending on Number of Degree2018

    • Author(s)
      Kyohei FUJII, Shuhei HASHIMOTO, Yoko UWATE and Yoshifumi NISHIO
    • Journal Title

      Proceedings of International SoC Design Conference

      Volume: - Pages: 164-165

    • Peer Reviewed
  • [Journal Article] Design of Convolutional Neural Network for Classifying Depth Prediction Images from Overhead2018

    • Author(s)
      Akari OURA, Kyohei FUJII, Yoko UWATE and Yoshifumi NISHIO
    • Journal Title

      Proceedings of International SoC Design Conference

      Volume: - Pages: 168-169

    • Peer Reviewed
  • [Presentation] Study on Improvement of Learning Ability of the Three-Layer Neural Networks2018

    • Author(s)
      Yuichi, MIYATA, Ryuta YOSHIMURA, Yoko UWATE and Yoshifumi NISHIO
    • Organizer
      International Workshop on Computer Vision and Signal Processing
    • Int'l Joint Research
  • [Presentation] Performance of Ant Colony Optimization Changing Characteristics of Pheromone's Reaction2018

    • Author(s)
      Ken KAMIYOTSUMOTO, Masaki TAKEUCHI, Yoko UWATE and Yoshifumi NISHIO
    • Organizer
      International Workshop on Computer Vision and Signal Processing
    • Int'l Joint Research
  • [Presentation] Effectiveness of Stacked Denoising Autoencoder with Chaotic Noise2018

    • Author(s)
      Ryuta YOSHIMURA, Yoko UWATE and Yoshifumi NISHIO
    • Organizer
      International Workshop on Computer Vision and Signal Processing
    • Int'l Joint Research
  • [Presentation] Synchronization Phenomena in a Ring of van der Pol Oscillators Coupled by Time-Varying Resistor2018

    • Author(s)
      Toshiya MATSUNASHI, Daiki NARIAI, Yoko UWATE and Yoshifumi NISHIO
    • Organizer
      International Workshop on Computer Vision and Signal Processing
    • Int'l Joint Research
  • [Presentation] Synchronization Phenomena in Coupled Chaotic Circuits with Asymmetric Weights Depending on Length2018

    • Author(s)
      Kyohei FUJII, Shuhei HASHIMOTO, Yoko UWATE and Yoshifumi NISHIO
    • Organizer
      International Workshop on Computer Vision and Signal Processing
    • Int'l Joint Research
  • [Presentation] Propagation of Chaotic and Periodic Solutions in Ladder Coupled Chaotic Circuits with Different Parameters2018

    • Author(s)
      Katsuya NAKABAI, Katsuki NAKASHIMA, Yoko UWATE and Yoshifumi NISHIO
    • Organizer
      International Workshop on Computer Vision and Signal Processing
    • Int'l Joint Research
  • [Presentation] Color Image Processing Method in Two Template Cellular Neural Networks2018

    • Author(s)
      Yasuteru HOSOKAWA, Yoko UWATE and Yoshifumi NISHIO
    • Organizer
      Taiwan and Japan Conference on Circuits and Systems
    • Int'l Joint Research
  • [Presentation] Investigation of Learning Accuracy in Cut Out Images Using Convolution Neural Network2018

    • Author(s)
      Yuichi MIYATA, Ryuta YOSHIMURA, Yoko UWATE and Yoshifumi NISHIO
    • Organizer
      Taiwan and Japan Conference on Circuits and Systems
    • Int'l Joint Research
  • [Presentation] Synchronization Phenomena in Coupled van der Pol Oscillators with Hourglass Structure2018

    • Author(s)
      Takumi NARA, Daiki NARIAI, Yoko UWATE and Yoshifumi NISHIO
    • Organizer
      Taiwan and Japan Conference on Circuits and Systems
    • Int'l Joint Research
  • [Presentation] Synchronization Phenomena in van der Pol Oscillators with Diodes2018

    • Author(s)
      Yuya IKEZOE, Toshiya MATSUNASHI, Yoko UWATE and Yoshifumi NISHIO
    • Organizer
      Taiwan and Japan Conference on Circuits and Systems
    • Int'l Joint Research
  • [Presentation] Neural Network Changed Characteristics of Rectified Linear Unit Function2018

    • Author(s)
      Toshiya MATSUNASHI, Kazuki NAGAO, Yoko UWATE and Yoshifumi NISHIO
    • Organizer
      Taiwan and Japan Conference on Circuits and Systems
    • Int'l Joint Research
  • [Presentation] Swarm Intelligence Algorithm Applying for Time-Varying Solutions2018

    • Author(s)
      Ken KAMIYOTSUMOTO, Thomas OTT, Yoko UWATE and Yoshifumi NISHIO
    • Organizer
      Taiwan and Japan Conference on Circuits and Systems
    • Int'l Joint Research
  • [Presentation] Classifying Overhead Images Predicted Depth by Fully Convolutional Residual Networks with Convolutional Neural Network2018

    • Author(s)
      Shu SUMIMOTO, Ryuta YOSHIMURA, Yoko UWATE and Yoshifumi NISHIO
    • Organizer
      Taiwan and Japan Conference on Circuits and Systems
    • Int'l Joint Research

URL: 

Published: 2019-12-27  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi