DOI QR코드

DOI QR Code

퍼지 제어 기법을 이용한 개선된 Max-Min 신경망

An Enhanced Max-Min Neural Network using a Fuzzy Control Method

  • 김광백 (신라대학교 컴퓨터공학과) ;
  • 우영운 (동의대학교 멀티미디어공학과)
  • 투고 : 2013.06.10
  • 심사 : 2013.08.23
  • 발행 : 2013.08.30

초록

본 논문에서는 퍼지 제어 기법을 적용하여 학습률을 자동으로 조정하는 개선된 Max-Min 신경망을 제안하였다. 개선된 Max-Min 신경망은 경쟁 단계에서 필요한 학습 시간을 줄이기 위하여, 정확성의 수와 부정확성의 수를 퍼지 제어 시스템의 입력으로 적용하여 학습률을 동적으로 조정하는 기법이다. 본 논문에서 제안된 방법을 실제 콘크리트 표면 균열 영상에서 추출한 균열의 방향성 패턴을 대상으로 인식 실험한 결과, 개선된 Max-Min 신경망이 효과적임을 확인할 수 있었다.

In this paper, we proposed an enhanced Max-Min neural network by auto-tuning of learning rate using fuzzy control method. For the reduction of training time required in the competition stage, the method was proposed that arbitrates dynamically the learning rate by applying the numbers of the accuracy and the inaccuracy to the input of the fuzzy control system. The experiments using real concrete crack images showed that the enhanced Max-Min neural network was effective in the recognition of direction of the extracted cracks.

키워드

참고문헌

  1. K. B. Kim, J. H. Cho, and E. Y. Cha, "A New Single Layer Perceptron using Fuzzy Neural Controller", Simulations International XII, Vol. 27, No. 7, pp. 341-343, 1995.
  2. K. B. Kim, C. K. Yuk and E. Y. Cha, "A New Fuzzy Supervised Learning Algorithm", AFSS'98, pp. 399-403, 1998.
  3. M. Delgado, I. Requena, "Solving fuzzy relational equations by max-min neural networks", Proceedings of IEEE World Congress on Computational Intelligence., Vol. 3, pp. 1737-1742, 1994.
  4. K. B. Kim, Y. W. Woo, " Detection of Flaws in Ceramic Materials Using Non-Destructive Testing", The Journal of The Korea Institute of Electronic Communication Sciences, Vol. 5, No. 3, pp. 321-326, 2010.
  5. K. B. Kim, M. H. Kang and E. Y. Cha, "A Fuzzy Competitive Backpropagation Using Nervous System", Proceedings of WCSS'97, pp. 188-193, 1997.
  6. K. B. Kim, W. J. Lee, Y. W. Woo, "Automatic Recognition and Performance of Printed Musical Sheets Using Fuzzy ART", The Journal of The Korea Institute of Electronic Communication Sciences, Vol. 6, No. 1 pp. 84-69, 2011.
  7. J. H. Cho, "Enhanced Fuzzy Binarization Method for Car License Plate Binarization", The Journal of The Korea Institute of Electronic Communication Sciences, Vol. 6, No. 2, pp. 231-236, 2011.
  8. K. B. Kim, H. J. Park,"Recognition of Concrete Surface Using Enhanced Max-Min Neural Networks", The Journal of The Korea Society of Computer and Information, Vol. 12, No. 2, pp. 77-82, 2007.

피인용 문헌

  1. Study on Map Matching Using Characteristics of Vehicular Movements vol.17, pp.5, 2015, https://doi.org/10.7855/IJHE.2015.17.5.075