Proceedings of the KIEE Conference (대한전기학회:학술대회논문집)

The Korean Institute of Electrical Engineers (대한전기학회)
권호 표지

The Information Diffusion Neural Networks for Real-Time Regeneration of 3-D Terrain Elevation Data with Contour Information

등고선 정보로부터 3차원 지형정보의 실시간 복원을 위한 정보 확산 신경회로망

  • Kim, J.M. (Dept. of Contol & Instrumention. Eng. Chonbuk National University) ;
  • Choi, J.S. (R&D Center, POSCON Co.) ;
  • Lim, Y.J. (Dept. of Contol & Instrumention. Eng. Chonbuk National University) ;
  • Kim, H.G. (Dept. of Contol & Instrumention. Eng. Chonbuk National University) ;
  • Kim, H.S. (Dept. of Contol & Instrumention. Eng. Chonbuk National University) ;
  • Kim, S.J. (Dept. of Contol & Instrumention. Eng. Chonbuk National University)
  • 김종만 (전북대학교 제어계측공학과) ;
  • 최종수 ((주)포스콘 연구소) ;
  • 임영재 (전북대학교 제어계측공학과) ;
  • 김홍갑 (전북대학교 제어계측공학과) ;
  • 김형석 (전북대학교 제어계측공학과) ;
  • 김성중 (전북대학교 제어계측공학과)
  • Published : 1995.07.20
Download PDF
⟨ Previous Next ⟩

Abstract

The Information Diffusion Neural Networks is proposed to regenerate the 3-dimensional terrain elevation data from contour lines. Contours in paper map are an expression of terrain elevation in highly compressed form. A real time regeneration of terrain data for each grid points from the the contour information is required for various applications. In the proposed neural networks, the elevation information on contours is diffused to neighbor units through updating its output toward that of neighbor units. An interpolation of terrain information is achieved from such computation mechanithm. Terrain data regeneration simulation has been done with sampled terrain data on contour lines.

Keywords