Fiber Recognition and Distribution Analysis of PET/Rayon Composite Yarn Cross Sections Using Image Processing Techniques

Abstract

Traditionally, the blending composition of a composite yams has been analyzed by human visual inspection, but such a method is time-consuming and inaccurate. In this study, we apply image processing techniques to analyzing cross sections of a PET/ Rayon composite yam, obtaining single fiber positions in the image with mask pro cessing. Fibers are recognized with four features: area, perimeter, compactness, and shape factor, and fiber distributions of cross sections of composite yarns are analyzed using three predefined index categories: the radial distribution index, the lateral distri bution index, and the angular distribution index.