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

Face Recognition System Using Codebook Space Information Processing

Research Project

Project/Area Number 11555090
Research Category

Grant-in-Aid for Scientific Research (B).

Allocation TypeSingle-year Grants
Section展開研究
Research Field 電子デバイス・機器工学
Research InstitutionTohoku University

Principal Investigator

KOTANI Koji  Tohoku University, Graduate School of Engineering, Associate Professor, 大学院・工学研究科, 助教授 (20250699)

Co-Investigator(Kenkyū-buntansha) OHMI Tadahiro  Tohoku University, New Industry Creation Hatchery Center, Professor, 未来科学技術共同研究センター, 教授 (20016463)
Project Period (FY) 1999 – 2000
Project Status Completed (Fiscal Year 2000)
Budget Amount *help
¥13,700,000 (Direct Cost: ¥13,700,000)
Fiscal Year 2000: ¥6,400,000 (Direct Cost: ¥6,400,000)
Fiscal Year 1999: ¥7,300,000 (Direct Cost: ¥7,300,000)
KeywordsFace Recognition / Codebook / Vector Quantization / Codebook Space Information / Facial Expression Recognition / Speaker Recognition / コードブックスペクトル
Research Abstract

Novel faco rccognition technology has been developed using "Vector-Quatization Codebook-Space Information Processing Algorithm." Detailed processing steps are as follows. Facial image is first divided into small size blocks (4X4) and differential intensity information is extracted by subtracting minimum intensity within the blocks. Then vector quantization is carried out using theoretically synthesized codebook and personal feature information is extracted by statistically analyzing referred frequenucies of each codebook vector. Finally, matching between the feature information and the database is carried out to identify the person. We have developed effective filtering procedure to eliminate noise and unwanted signal component and histogram standardization procedure to calibrate the size of faces. In addition, we have introduced "effective discrimination distanace" as a recognition measture in order to improve the recognition algorithm at higher success rate region. Finally, we have r … More ealized 100% recognition success rate for 44 person's 220 facial images.
We have applied the recognition algorithm into the facial expression recognition. 100% success rate has been realized in recognizing 3 facial expressions (anger, happiness, and normal) of the identical person. It is revealed by evaluating suitable filter size and image resolution that the signal components having a period of 13mm to 14 mm or longer at real space are very important for facial expression recognition.
We have also studied a speaker recognition technology for realizing highly accurate human recognition and identification. We have realized high performance speaker recognition algorithm, which utilizes feature extraction by cepstrum analysis and classification by vector quantization. Improvements on recognition speed and success rate were achieved by newly developed hierarchical matching method and pre-learning procedure. Finally, we have realized 97% recognition success rate at maximum for 58 person's 290 text-independent speeches. Less

Report

(3 results)
  • 2000 Annual Research Report   Final Research Report Summary
  • 1999 Annual Research Report
  • Research Products

    (26 results)

All Other

All Publications (26 results)

  • [Publications] Z.Pan,K.Kotani,T.Ohmi: "Extracting person's speech individually from original records of meeting by speaker identification technique"Technical Report of IEICE. SP99-77. 9-13 (1999)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2000 Final Research Report Summary
  • [Publications] Z.Pan,K.Kotani,T.Ohmi: "A system for generating speech index based on speaker identification technique through VQ"The 3rd workshop on system LSI, Biwako. 235-238 (1999)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2000 Final Research Report Summary
  • [Publications] Z.Pan,K.Kotani,T.Ohmi: "A nonlinear cepstral compensation method for noisy speech processing"Technical Report of IEICE. SP99-103. 61-65 (1999)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2000 Final Research Report Summary
  • [Publications] Z.Pan,K.Kotani,T.Ohmi: "A speech indexing system for recorded audio source based on speaker recognition technique"International conference on advances in intelligent systems. 63-64 (2000)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2000 Final Research Report Summary
  • [Publications] Z.Pan,K.Kotani,T.Ohmi: "A On-Line Hierarchical Method of Speaker Identification for Large Population"Nordic Signal Processing Symposium. 33-35 (2000)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2000 Final Research Report Summary
  • [Publications] Z.Pan,K.Kotani,T.Ohmi: "A speech indexing system for recorded audio source based on speaker identification technique"Advances in Intelligent Systems : Theory and Applications, Edited by Masoud Mohamadian, IOS Press, Ohmsha. 239-243 (2000)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2000 Final Research Report Summary
  • [Publications] Z.Pan,K.Kotani,T.Ohmi: "A novel method of speaker identification for large population by pre-learning of test utterance using vector quantization"Proceedings, World Multiconference on Systemics, Cybernetics and Informatics. Vol.VI,Part II. 48-51 (2000)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2000 Final Research Report Summary
  • [Publications] Z.Pan,K.Kotani,T.Ohmi: "A Fast Search Method of Speaker Identification for Large Population Using Pre-selection and Hierarchical Matching"6th International Conference on Spoken Language Processing. 290-293 (2000)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2000 Final Research Report Summary
  • [Publications] Z.Pan,K.Kotani,T.Ohmi: "A Fast Search Method of VQ-Based Speaker Identification for Large Population Using Discriminative Factor and Hierarchical Matching"2001 International Conference on Acoustics, Speech and Signal Processing (ICASSP) . (Now Printing). (2001)

    • Description
      「研究成果報告書概要(和文)」より
    • Related Report
      2000 Final Research Report Summary
  • [Publications] Zhibin Pan, Koji Kotani, Tadahiro Ohmi: "Extracting person's speech individually from original records of meeting by speaker identification technique."Technical Report of IEICE. Vol.99, No.298, DSP99-80, SP99-78. 9-13 (1999)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2000 Final Research Report Summary
  • [Publications] Zhibin Pan, Koji Kotani, Tadahiro Ohmi: "A system for generating speech index based on speaker identification technique through VQ."The 3rd workshop on system LSI, Biwako. 235-238 (1999)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2000 Final Research Report Summary
  • [Publications] Zhibin Pan, Koji Kotani, Tadahiro Ohmi: "A nonlinear cepstral compensation method for noisy speech processing"Technical Report of IEICE. NLC99-35, SP99-103. 61-65 (1999)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2000 Final Research Report Summary
  • [Publications] Zhibin Pan, Koji Kotani, Tadahiro Ohmi: "A speech indexing system for recorded audio source based on speaker recognition technique."International Conference on Advances in Intelligent Systems : Theory and Applications Canberra. 63-64 (2000)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2000 Final Research Report Summary
  • [Publications] Zhibin Pan, Koji Kotani, Tadahiro Ohmi: "A Oh-Line Hierarchical Method of Speaker Identification for Large Population."Nordic Signal Processing Symposium Sweden. 33-35 (2000)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2000 Final Research Report Summary
  • [Publications] Zhibin Pan, Koji Kotani, Tadahiro Ohmi: "A speech indexing system for recorded audio source based on speaker identification technique."Advances in Intelligent Systems : Theory and Applications, Edited by Masoud Mohamadian, IOS Press, Ohmsha. 239-243 (2000)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2000 Final Research Report Summary
  • [Publications] Zhibin Pan, Koji Kotani, Tadahiro Ohmi: "A novel method of speaker identification for large population by pre-learning of test utterance using vector quantization."Proceedings, World Multiconference on Systemics, Cybernetics and Informatics Orland. Vol.VI, Part II. 48-51 (2000)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2000 Final Research Report Summary
  • [Publications] Zhibin Pan, Koji Kotani, Tadahiro Ohmi: "A Fast Search Method of Speaker Identification for Large Population Using Pre-Selectrion and Hierarchical Matching."6th International Conference on Spoken Language Processing China. 290-293 (2000)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2000 Final Research Report Summary
  • [Publications] Zhibin Pan, Koji Kotani, Tadahiro Ohmi: "A Fast Search Method of VQ-BAsed Speaker Identification for Large Population Using Discriminative Factor and Hierarchical Matching."2001 International Conference on Acoustics, Speech and Signal Processing (ICASSP). (Now Printing). (2001)

    • Description
      「研究成果報告書概要(欧文)」より
    • Related Report
      2000 Final Research Report Summary
  • [Publications] Z.Pan,K.Kotani,T.Ohmi: "A On-Line Hierarchical Method of Speaker Identification for Large Population"Nordic Signal Processing Symposium. 33-35 (2000)

    • Related Report
      2000 Annual Research Report
  • [Publications] Z.Pan,K.Kotani,T.Ohmi: "A speech indexing system for recorded audio source based on speaker identification technique"Advances in Intelligent Systems : Theory and Applications, Edited by Masoud Mohamadian, IOS Press, Ohmsha. 239-243 (2000)

    • Related Report
      2000 Annual Research Report
  • [Publications] Z.Pan,K.Kotani,T.Ohmi: "A novel method of speaker identification for large population by pre-learning of test utterance using vector quantization"Proceedings, World Multiconference on Systemics, Cybernetics and Informatics. Vol.VI,Part II. 248-251 (2000)

    • Related Report
      2000 Annual Research Report
  • [Publications] Z.Pan,K.Kotani,T.Ohmi: "A Fast Search Method of Speaker Identification for Large Population Using Pre-selection and Hierarchical Matching"6th International Conference on Spoken Language Processing. 290-293 (2000)

    • Related Report
      2000 Annual Research Report
  • [Publications] Z.B.Pan: "Extracting person's speech individually from original records of meeting by speaker identification technique"Technical Report of IEICE. DSP99-79SP99-77. 9-13 (1999)

    • Related Report
      1999 Annual Research Report
  • [Publications] Z.B.Pan: "A system for generating speech index based on speaker identincation technique through VQ"The 3rd workshop on system LSI, Biwako. 235-238 (1999)

    • Related Report
      1999 Annual Research Report
  • [Publications] Z.B.Pan: "A nonlinear cepstral compensation method for noisy speech processing"The 1st symposium on speech and language processing. (1999)

    • Related Report
      1999 Annual Research Report
  • [Publications] Z.B.Pan: "A speech indexing system for recorded audio source based on speaker recognition technique"International conference on advances in intelligent systems. 63-64 (2000)

    • Related Report
      1999 Annual Research Report

URL: 

Published: 1999-04-01   Modified: 2016-04-21  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi