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

Discovery of structure from color images by entropy minimization learning of neural networks

Research Project

Project/Area Number 10480076
Research Category

Grant-in-Aid for Scientific Research (B)

Allocation TypeSingle-year Grants
Section一般
Research Field Intelligent informatics
Research InstitutionKyushu University

Principal Investigator

NIJIMA Koichi  Kyushu University, Graduate School of Information Science and Electrical Engineering, Professor, 大学院・システム情報科学研究院, 教授 (30047881)

Co-Investigator(Kenkyū-buntansha) TAKANO Shigiru  Kyushu University, Graduate School of Information Science and Electrical Engineering, Assistant Professor, 大学院・システム情報科学研究院, 助手 (70336064)
TAKAHASHI Norikazu  Kyushu University, Graduate School of Information Science and Electrical Engineering, Associate Professor, 大学院・システム情報科学研究院, 助教授 (60284551)
OKADA Yoshihiro  Kyushu University, Graduate School of Information Science and Electrical, Engineering Associate Professor, 大学院・システム情報科学研究院, 助教授 (70250488)
葛目 幸一  弓削商船高等学校, 情報工学科, 助教授 (80225151)
皆本 晃弥  九州大学, 大学院・システム情報科学研究科, 助手 (00294900)
ツォイクマン トーマス  九州大学, 大学院・システム情報科学研究科, 教授 (60264016)
Project Period (FY) 1998 – 2001
Project Status Completed (Fiscal Year 2001)
Budget Amount *help
¥10,300,000 (Direct Cost: ¥10,300,000)
Fiscal Year 2001: ¥2,400,000 (Direct Cost: ¥2,400,000)
Fiscal Year 2000: ¥2,500,000 (Direct Cost: ¥2,500,000)
Fiscal Year 1999: ¥2,400,000 (Direct Cost: ¥2,400,000)
Fiscal Year 1998: ¥3,000,000 (Direct Cost: ¥3,000,000)
KeywordsNeural network / Structure of image / Image classification / Image recognition / Rule extraction / Entropy minimization / Surface generation / 類似検索 / 訓練パターン / 認識領域 / 枝刈り / 曲面生成 / マージ法 / 位相 / 境界処理 / 教師付きエントロピー最大化法 / 自己組織型エントロピー最大化法 / データマイニング / 隠蔽画像 / ウェーブレット変換 / 画像作成システム / ウェーブレット / 学習理論 / 安定性理論 / 動画像認識 / 画像処理 / 非対称相互結合型 / 引き込み領域
Research Abstract

Neural networks play an important role in this research. In order to discover the structure that characterizes images, we need to study learning methods for neural networks and to analyze the behavior of the learned neural network. It is required to study the classification and recognition ability of the neural network. We must find rules from the learned neural network. Simplification of 3D images for inputting in the neural network is also an important research subject.
To resolve such problems, we studied various learning methods as well as the entropy minimization learning technique for neural networks and the behavior of cellular neural networks. And we applied the obtained results to image classification and rule extraction. We proposed a method for finding a hidden image from two similar images by combining the entropy minimization learning technique of the neural network with a wavelet theory, which is a main theme of this research. For discovering the structure of 3D models using a neural network, we studied simple surface generation algorithms for obtaining vertices which are input data of the neural network.

Report

(5 results)
  • 2001 Annual Research Report   Final Research Report Summary
  • 2000 Annual Research Report
  • 1999 Annual Research Report
  • 1998 Annual Research Report
  • Research Products

    (31 results)

All Other

All Publications (31 results)

  • [Publications] K.Niijima et al.: "Convergence rate of minimization learning for neural networks"Lecture Note in Artificial Intelligence. 1398. 412-417 (1998)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2001 Final Research Report Summary
  • [Publications] A.Ohkubo, K.Niijima: "A new supervised learning method of neural networks and its application to the land cover classification"Proc. of IEEE International Geoscience and Remote Sensing symposium. 1369-1371 (1999)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2001 Final Research Report Summary
  • [Publications] A.Ohkubo, K.Niijima: "New supervised learning of neural networks for satellite for image classification"Proc. of IEEE International Conference on Image Processing. 505-509 (1999)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2001 Final Research Report Summary
  • [Publications] M.H.Mohamed, K.Niijima: "Extracting rules from neural networks by removing unnecess ary connections"Proc. of the Second ICSC Symposium on Neural Computation. 322-328 (2000)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2001 Final Research Report Summary
  • [Publications] K.Niijima et al.: "Efficient simplification of polygonal surface models"Proc. of 5th International Conference on Information Visualization. 464-469 (2001)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2001 Final Research Report Summary
  • [Publications] K.Niijima et al.: "Fast, simple and memory efficient mesh simplification"Proc. of 4^<th> IASTED International Conference on Computer Graphics and Imaging. 72-75 (2001)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2001 Final Research Report Summary
  • [Publications] 新島 耕一: "発見科学とデータマイニング(bit別冊,分担執筆)"共立出版. 300 (2000)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2001 Final Research Report Summary
  • [Publications] M.H. Mohamed, T. Minamoto and K. Niijima: "Convergence rate of minimization learning for neural networks"Proc. of the Tenth European Conference on Machine Learning, Vol.1398, Lecture Note in Artificial Intelligence. 412-417 (1998)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2001 Final Research Report Summary
  • [Publications] A. Ohkubo and K. Niijima: "A new supervised learning method of neural networks and its application to the land cover classification"Proc. of IEEE International Geoscience and Remote Sensing Symposium. 1369-1371 (1999)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2001 Final Research Report Summary
  • [Publications] A. Ohkubo and K. Niijima: "New supervised learning of neural networks for satellite for image classification"Proc. of IEEE International Conference on Image Processing. 505-509 (1999)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2001 Final Research Report Summary
  • [Publications] M. H. Mohamed and K. Niijima: "Extracting rules from neural networks by removing unnecessary connections"Proc. of the Second ICSC Symposium on Neural Computation. 322-328 (2000)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2001 Final Research Report Summary
  • [Publications] M. Hussain, Y. Okada and K. Niijima: "Efficient simplification of polygonal surface models"Proc. of 5th International Conference on Information Visualization. 464-469 (2001)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2001 Final Research Report Summary
  • [Publications] M. Hussain, Y. Okada and K. Niijima: "Fast, simple and memory efficient mesh simplification"Proc, of the Fourth IASTED International Conference on Computer Graphics and Imaging. 72-77 (2001)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2001 Final Research Report Summary
  • [Publications] K.Niijima, M.H.Mohamed: "Neural networks as domains of classification"Technical Report of Department of Informatics. No.186. 1-15 (2001)

    • Related Report
      2001 Annual Research Report
  • [Publications] M.Hussain, Y.Okada, K.Niijima: "Efficient simplification of polygonal surface models"Proceedings of Fifth International Conference on information Visualization (IV2001), IEEE Computer Society Press. 464-469 (2001)

    • Related Report
      2001 Annual Research Report
  • [Publications] M.Hussain, Y.Okada, K.Niijima: "Fast, simple and memory efficient mesh simplification"Proceedings of the Fourth IASTED International Conference on Computer Graphics and Imaging (CGIM2001), ACTA Press. 72-77 (2001)

    • Related Report
      2001 Annual Research Report
  • [Publications] M.H..Mohamed and K.Niijima: "Extracting rules from neural networks by removing unnecessary connections"Proc.ICSC Symposium on Neural Computation. (CD-ROM). (2000)

    • Related Report
      2000 Annual Research Report
  • [Publications] M.H.Mohamed and K.Niijima: "Redundant connection effect on the error rate for the neural networks"Proc.Seventh International Conference on Neural Information Processing. 2. 981-985 (2000)

    • Related Report
      2000 Annual Research Report
  • [Publications] Y.Okada,E.Itoh and S.Hirokawa: "IntelligentBox : Its aspect as a rapid construction system for interactive 3D games"Proc.First International Conference on Intelligent Games and Simulation. 1. 22-26 (2000)

    • Related Report
      2000 Annual Research Report
  • [Publications] 新島耕一: "発見科学とデータマイニング(分担)"共立出版. 201-206 (2000)

    • Related Report
      2000 Annual Research Report
  • [Publications] Akito Ohkubo and Koichi Niijima: "A new supervised learning method of neural networks and its application to the land cover classfication"Proc.IEEE Int.Geoscience and Remote Sensing Sympo. 1369-1371 (1999)

    • Related Report
      1999 Annual Research Report
  • [Publications] Akito Ohkubo and Koichi Niijima: "New supersed learning of neural networks for satellite image classification"Proc.IEEE Int.Conf.on Image Processing. 505-509 (1999)

    • Related Report
      1999 Annual Research Report
  • [Publications] Marghny Hassan Mohamed and Koichi Niijima: "Rules extraction by constructive learning of neural networks and hidden-unit clustering"Proc.2nd Int.Conf.on Discovery Scinece,Lecture Notes in Artificial Intelligence. 1721. 343-344 (1999)

    • Related Report
      1999 Annual Research Report
  • [Publications] Norikazu Takahashi: "Stability analysis of cellular neural networks with delay"Proc.1999 Int.Sympo.on Nonlinear Theory and its Applications. 403-406 (1999)

    • Related Report
      1999 Annual Research Report
  • [Publications] Frank Stephan and Thomas Zeugmann: "On the uniform learnability of approximations to non-recursive functions"Proc.10th Int.Conf.Algorithmic Learning Theory,Lecture Notes in Artificial Intelligence. 1720. 276-290 (1999)

    • Related Report
      1999 Annual Research Report
  • [Publications] M.H.Mohamed, T.Minamoto and Koichi Niijima: "Convergence Rate of Minimization Learning for Neural Networks" Machine Learning:ECML-98, Lecture Notes in Artificial Intelligence. 1398. 412-417 (1998)

    • Related Report
      1998 Annual Research Report
  • [Publications] S.Takano, T.Minamoto and K.Niijima: "Moving Object Recognition Using Wavelets and Learning of Eigenspaces" Discovery Science, Lecture Notes in Artificail Intelligence. 1532. 443-444 (1998)

    • Related Report
      1998 Annual Research Report
  • [Publications] M.H.Mohamed, A.Okubo, T.Minamoto and K.Niijima: "Learning of Three-Layer Neural Networks Based on Domains of Attraction" Proc.Fifth Int.Conf.Neural Information Processing. 1591-1594 (1998)

    • Related Report
      1998 Annual Research Report
  • [Publications] K.Kuzume and K.Niijima: "Design of Optimal Lifting Wavelet Filters for Data Compression" Proc.IEEE-SP TF/TS Int.Sympo.337-340 (1998)

    • Related Report
      1998 Annual Research Report
  • [Publications] N.Takahashi and L.O.Chua: "On the Complete Stability of Nonsymmetric Cellular Neural Networks" IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I. 45・7. 754-758 (1998)

    • Related Report
      1998 Annual Research Report
  • [Publications] R.Reischuk and T.Zeugmann: "Learning One-variable Pattern Languages in Linear Average Time" Proceedings of the Eleventhe Annual Conference on Computational Learning Theory. 198-208 (1998)

    • Related Report
      1998 Annual Research Report

URL: 

Published: 1998-04-01   Modified: 2016-04-21  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi