Determining Fiber Orientation Distribution in Nonwovens with Hough Transform Techniques

Abstract

This research explores the application of an image transform technique, the Hough transform (ht), to measuring fiber orientation distribution in nonwoven fabrics. In this paper, various forms of the ht for straight lines and the implementing algorithms are presented, the factors that may influence the accuracy of ht results are discussed, and the way to determine fiber orientation distribution from ht data for a nonwoven is explained. The usefulness of the ht method is tested using simulated nonwoven images with known dominant fiber orientations, and nonwoven fabrics differing in fiber content, weight, and other characteristics. Fiber orientation distributions generated by the ht method are also compared with those obtained from the zero-span tensile testing method. The principal advantages of using the ht are that it is robust in suppressing image noise and is able to deal with fibers containing gaps and breaks. Measurements with the ht method show good correlation with those obtained from other methods. The ht appears to be an effective and efficient image analysis technique for extracting information about fiber orientation from images of nonwoven fabrics.