Expression of individual woven yarn of textile fabric based on segmentation of three dimensional CT image by using yarn position and orientation

In this paper, an expression method of the individual woven yarn of a textile fabric is proposed for structure analysis of textile fabric on its three-dimensional (3D) X-ray computed tomography (CT) image. In this method, the voxels of the 3D CT image is segmented into each yarn based on the yarn position, which is the position of the sequence of the yarn center points, and the yarn orientation. First of all, the orientations of the filaments at all the voxels of the 3D CT image are estimated by correlating the voxel values with a filament model. Each filament is then reconstructed by connecting the voxels of the filaments which have similar estimated filament orientations. Each yarn is reconstructed by bundling the reconstructed filaments based on distance between two filaments. The yarn position and orientation are obtained by thinning the reconstructed yarns. The voxels of the 3D CT image are segmented into each yarn by clustering on the basis of newly defined distance between the voxel and the yarn center point. As a result, the individual yarn is expressed by volume-rendering technique with the segmented voxels. The effectiveness of the proposed yarn expression method is confirmed by experimentally applying this method to a 3D image of a plain knitted fabric.