Algorithm 828: DNSPLIN1: discrete nonlinear spline interpolation

  • Authors:
  • Robert J. Renka

  • Affiliations:
  • University of North Texas, Denton, Texas

  • Venue:
  • ACM Transactions on Mathematical Software (TOMS)
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

We describe a new method and a Fortran-77 code for constructing discrete approximations to nonparametric interpolating nonlinear spline curves. Our approach consists of minimizing the discretized strain energy by a descent method with a Sobolev gradient in place of the standard gradient. It serves as a demonstration of the Sobolev gradient method, which is much more generally applicable. The effectiveness of the method in rapidly producing smooth interpolatory curves is demonstrated by test results for several challenging data sets.