Image processing and data evaluation algorithms for reproducible optical in-plane permeability characterization by radial flow experiments

Abstract

Textile permeability is one of the dictating factors in the fabrication of fibre-reinforced polymer composites. However, reproducibility of experimental in-plane permeability characterization is still a challenging task due to the lack of standardized test and evaluation procedures. The paper at hand addresses two major sources for discrepancies when characterizing in-plane permeability through optical observation of radial flow experiments: digital image processing and data evaluation algorithms. A digital image processing strategy is presented, which robustly handles varying lightning conditions, optical properties of the materials under test and image occlusions caused by mechanical elements of the test setup. The strategy is of universal validity and independent of the choice of reinforcing material and impregnating fluid. An experimental analysis compares two approaches for fitting elliptic geometry models to data points detected along the fluid flow front. The study reveals the impact of the fitting strategy on the resulting permeability data and the benefit of forcing the ellipse centre to that of the injection opening. The computation algorithm of Chan and Hwang, widely used for calculating in-plane permeability values from experimental data, is critically discussed. A correction of the algorithm is proposed which avoids a violation of isotropic data characteristics while adding robustness to the data reduction. An experimental analysis compares anisotropic in-plane permeability values obtained with different evaluation algorithms. The study highlights the impact of the computational algorithm on the permeability data and reveals discrepancies of up to 6%, which is considerable compared to the scatter typically reported for in-plane permeability data.