KINDAI UNIVERSITY


*A space between the first name and last name, please enter

TAKEDA Fumiaki

Profile

FacultyDepartment of Electronic Engineering and Computer Science / Graduate School of System Enginnering / Research Institute of Fundamental Technology for Next Generation
PositionProfessor
Degree
Commentator Guidehttps://www.kindai.ac.jp/meikan/1418-takeda-fumiaki.html
URL
Mail
Last Updated :2020/09/30

Research Activities

Research Areas

  • Manufacturing technology (mechanical, electrical/electronic, chemical engineering), Control and systems engineering
  • Manufacturing technology (mechanical, electrical/electronic, chemical engineering), Control and systems engineering
  • Manufacturing technology (mechanical, electrical/electronic, chemical engineering), Control and systems engineering
  • Manufacturing technology (mechanical, electrical/electronic, chemical engineering), Control and systems engineering

Research Interests

  • System Engineering

Published Papers

  • Generation Method of the Trigger Signal for a Full Automatic Capture Device to the Harmful Animals with Color and Infrared Depth Image, TAKEDA Fumiaki, 18(1), 21 - 32, Dec. 2016 , Refereed
  • Carved seal recognition method of the steel spherical surface with mirror reflection using RBF typed neural network, TAKEDA Fumiaki, Transactions of the JSME, Transactions of the JSME, 82(835), 1 - 11, Mar. 2016 , Refereed
  • Analysis on the robustness of the pressure-based individual identification system based on neural networks, Lina Mi, Fumiaki Takeda, INTERNATIONAL JOURNAL OF INNOVATIVE COMPUTING INFORMATION AND CONTROL, INTERNATIONAL JOURNAL OF INNOVATIVE COMPUTING INFORMATION AND CONTROL, 3(1), 97 - 110, Feb. 2007 , Refereed
    Summary:In this paper, an individual identification system based on dynamic feature of signature pressure is introduced first. Then uniformed characters are proposed as register characters for the system, instead of traditionally employed individual signature,to address the problem of great difference in the recognition capabilities of the system for different registrants. To evaluate the effectiveness of the proposed method, twenty people are selected as target registrants and the recognition capabilities of the original signature-based systems are studied and compared with that of the uniformed-register-characters-based systems in experiment section. The results show that with the uniformed register characters, the proposed system seems to have important merit on the stability in terms of recognition capabilities for different registrants.
  • Alarm sound classification system of oxygen concentrator by using neural network, Fumiaki Takeda, Yuhki Shiraishi, Takeo Sanechika, INTERNATIONAL JOURNAL OF INNOVATIVE COMPUTING INFORMATION AND CONTROL, INTERNATIONAL JOURNAL OF INNOVATIVE COMPUTING INFORMATION AND CONTROL, 3(1), 211 - 222, Feb. 2007 , Refereed
    Summary:Domiciliary oxygen therapy uses oxygen concentrators, which sound alarms in the event of trouble, and the kind of alarm depends on that of the trouble. To lessen the burden on caregivers, we have proposed a system classifying alarm sounds using a multilayer perceptron neural network and transmitting the recognized trouble to a monitor center in a hospital by phone cable. This system has been constructed by a digital signal processor board as an external unit. The difference among kinds of alarm sounds are characterized by the length of alarm sounds and that of alarm periods. Therefore, a short-time Fourier transform is used as a feature extraction method. Experiments using real-world devices indicate that the proposed system works well under noisy conditions.

Works

  • Research and Development for Intelligent Systems, TAKEDA Fumiaki, Software
  • Research and Development for Intelligent Inspection Systems, TAKEDA Fumiaki, Software
  • Function Generation Systems by Learning of the Neural Network, TAKEDA Fumiaki, Software
  • Development of a arm behavior recognition system using electromyogram
  • Development of a measuring system for foods taking volume
  • Improvement of currency recognition system with neural network

Misc

  • Development of Neuro-Templates Matching Recognition Method for Banknotes, Proceedings of SCORE, 1,   2001
  • Development of Self-Learning Neuro-Recognition Board for Banknotes and its Wide Use Expansion, Trans. IEE of Japan, 121-C, 2, 357, 365,   2001
  • Development of a Neuro-Templates Matching Recognition Method for Banknotes, Trans. IEE of Japan, 121-C, 1, 196, 205,   2001
  • Development of Autonomic Neural Board and Advancement to Palm Prints Recognition, Proceedings of TENCON, 2, 160, 164,   2000
  • Multiple kinds of Paper Currency Recognition using Neural Network and application for Euro Currency, Proceedings of IEEE-INNS-ENNS IJCNN, 1,   2000
  • A Proposal of Structure Method for Multi-Currency Simultaneous Recognition using Neural Networks, Fumiaki Takeda, Toshihiro Nishikage, Nihon Kikai Gakkai Ronbunshu, C Hen/Transactions of the Japan Society of Mechanical Engineers, Part C, 66, 648, 2653, 2659,   2000 , 10.1299/kikaic.66.2653
    Summary:Up to now, we have developed banking machines for multi-currency recognition using neural networks (NNs). In this paper, we expand neuro-recognition system to increase the more number of recognition patterns using two image sensors, one sensor's purpose is discrimination for known image and another one is exclusion for unknown image. Concretely, we implement the proposed method to an experimental system, which has two sensors. And they are arranged on the up side and down side of the aisle, respectively. Finally, we show the effectiveness of the proposed method using this system, numerically. © 2000, The Japan Society of Mechanical Engineers. All rights reserved.
  • Banknote recognition by means of optimized masks, neural networks and genetic algorithms, F Takeda, T Nishikage, S Omatu, ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 12, 2, 175, 184,   1999 04 , 10.1016/S0952-1976(98)00061-X
    Summary:Previous work by the authors has proposed a banknote recognition system using a neural network (NN) to develop new types of banknote recognition machines. This system is constructed by means of some core techniques. One is a small-scale neural recognition technique using masks. The second is a mask-optimization technique using a genetic algorithm (GA), The last is a neural hardware technique using a digital signal processor (DSP). This paper focuses on and discusses the mask optimization by the GA, which is the second core technique in the neural recognition system. This technique enables the selection of good masks, that can effectively generate the characteristic values of the input image. Further, the effectiveness of this technique is shown not only by the generalization of the NN, but also by a statistical analysis, using the Italian banknotes. Finally, the feasibility and effectiveness of the neural recognition system is shown by using worldwide banknotes. (C) 1999 Elsevier Science Ltd. All rights reserved.
  • Neural Network Systems Technique and Applications in Paper Currency Recognition, Neural Network Systems, Techniques and Applications ACADEMIC Press, 5, 133, 160,   1998
  • A Neuro-Money Recognition Using Optimized Masks by GA, Advances in Fuzzy Logic, Neural Networks and Genetic Algorithms LNAI 1011, 190, 201,   1995
  • HIGH-SPEED PAPER CURRENCY RECOGNITION BY NEURAL NETWORKS, F TAKEDA, S OMATU, IEEE TRANSACTIONS ON NEURAL NETWORKS, 6, 1, 73, 77,   1995 01 , 10.1109/72.363448
    Summary:In this paper a new technique is proposed to improve the recognition ability and the transaction speed to classify the Japanese and U.S. paper currency. Two types of data sets, time series data and Fourier power spectra, are used in this study. In both cases, they are directly used as inputs to the neural network. Still more we also refer a new evaluation method of recognition ability. Meanwhile, a technique is proposed to reduce the input scale of the neural network without preventing the growth of recognition. This technique uses only a subset of the original data set which is obtained using random masks. The recognition ability of using large data set and a reduced data set are discussed. In addition to that the results of using a reduced data set of the Fourier power spectra and the time series data are compared.

Research Grants & Projects

  • Developement of a individual recognition system using biometrics on the Internet
  • Development of weake up sensing system of human behavior
  • Development of a intelligent recognition board using digital signal processor
  • Research of individual recognition system with biometrics and its development of prototype
  • Research of artificial intelligent currency recognition method by autonomic learning and its implementation to real system.