Proceedings of the Korean Institute of Intelligent Systems Conference (한국지능시스템학회:학술대회논문집)

Korean Institute of Intelligent Systems (한국지능시스템학회)
권호 표지

A Novel Fuzzy Morphology, Part II:Neural Network Implementation

  • Yonggwan Won (Electronics and Telecommunications Research Institute Taejon-City, Yooosung-ku, Yoosung, Kajung-Dong 161) ;
  • Lee, Bae-Ho (Department of Computer Engineering Chonnam National University Kwangju-City, Puk-ku, Yongbong-Dong 300)
  • Published : 1995.10.01
Download PDF
⟨ Previous Next ⟩

Abstract

A shared-weight neural network that performed classification based on the features extracted with the fuzzy morphological operation is introduced. Learning rules for the structuring elements, degree of membership, and weighting factors are also precisely described. In application to handwritten digit recognition problem, the fuzzy morphological shared-weight neural network produced the results which are comparable to the state-of-art for this problem.

Keywords