Efficient Parametric Curved Glass Fiber Representations

Tim Elberfeld, Bernhard Fröhler, Monica J. Emerson, Christoph Heinzl, Johann Kastner, Jan De Beenhouwer, Jan Sijbers

Research output: Contribution to conferencePaper


Glass fiber-reinforced polymers (GFRP) are highly versatile material components that are used in a wide variety of applications. Knowing the geometries of the fibers inside such a sample is important to estimate the mechanical properties of the respective samples. We present an approach for representing fiber traces in a sparse way using arc length parametrization. Two arc length representations based on polynomials and cosine basis functions are compared. The resulting curves are evaluated in terms of their deviation from high resolution fiber traces of simulated and real data. Furthermore, the sensitivity of the parameters to small permutations is investigated to estimate their suitability in numerical optimization within our existing fiber analysis framework.
Original languageEnglish
Number of pages7
Publication statusAccepted/In press - 2021
Event20th World Congress of Non-Destructive Testing (WCNDT) - Incheon, Korea, Republic of
Duration: 31 May 20214 Jun 2021


Conference20th World Congress of Non-Destructive Testing (WCNDT)
CountryKorea, Republic of
Internet address


  • Visualisierung
  • Visual Computing

Fingerprint Dive into the research topics of 'Efficient Parametric Curved Glass Fiber Representations'. Together they form a unique fingerprint.

Cite this