Measuring Fiber Orientation in Nonwovens

Abstract

In nonwovens, fiber orientation is an important characteristic because it directly influences the properties of the material. In a series of papers, we plan to explore the viability of image analysis algorithms for characterizing fiber orientation. In the first part of this series, we present a simulation scheme to produce images to test these different algorithms. As part of the series, we will present a direct tracking method in Part II, as well as Fourier and flow field analyses in subsequent parts for characterizing fiber orientation.

Distance transform algorithm for measuring nanofiber diameter

“…These are called S-randomness, µ-randomness (suitable for generating a web of continuous filaments) and I-randomness (suitable for generating a web of staple fibers). These methods have been described in elaborations by Abdel-Ghani and Davis [23] and Pourdeyhimi et al [9]. Since the physical characteristics of simulations are known exactly, one can employ them to test the usefulness of the algorithm used in characterizing diameter and other structural features [9].…”

“…These methods have been described in elaborations by Abdel-Ghani and Davis [23] and Pourdeyhimi et al [9]. Since the physical characteristics of simulations are known exactly, one can employ them to test the usefulness of the algorithm used in characterizing diameter and other structural features [9]. In this study, µ-randomness procedure has been used for generating simulated images with known characteristics.…”

“…In the last few years, image analysis methods have been developed in order to identify fibers and measure nonwoven characteristics such as fiber orientation [9][10][11][12][13][14][15][16], fiber diameter [17,18], pore size [19], [20], uniformity [21] and other structural features [15,22]. However, since these are new techniques and their accuracy and limitations have not been verified, samples with known characteristics are required to evaluate the accuracy of the methods which can be produced by simulation schemes [9,23].…”

2D finite element analysis of thermally bonded nonwoven materials: Continuous and discontinuous models

“…Therefore, the present model assumes the fibres are straight and arranged according to orientation distributions. The orientation distribution of the studied fabric was measured at North Carolina State University [11][12][13]. To generate the random fibrous network, a program was developed using Python language to describe the random fibrous assembly according to their orientation distribution.…”

“…This theory is based on the incremental deformation principle and predicts the material behaviour with the consideration of fibre properties, their orientation distribution function (ODF) and bonding deformation [8][9][10][11][12][13][14][15][16].…”