Measuring Structural Characteristics of Fiber Segments in Nonwoven Fabrics

Abstract

This paper describes the image analysis techniques used for measuring structural characteristics of fiber or fiber bundle segments in a thin nonwoven fabric. The mea surable characteristics include length, thickness, curl, and orientation of fiber segments. A new skeletonization algorithm for extracting the simplified structure from a non woven image is developed. Thickness of fiber segments is calculated directly in the process of skeletonizing the web, while the length, orientation, and curl of fiber segments are measured in the skeletonized image. Fiber segments are identified and measured by tracing skeletons between any two adjacent intersecting points. For a curled segment, the best-fitting line is used to calculate its orientation angle. Several nonwoven fabrics are tested, and the results are compared with those measured by other methods. This image system enables the users to obtain comprehensive measurements on fiber seg ments in a single procedure.