Abstract:

To recognize the weave of woven fabrics, the crossing points should first be located and recognized. To overcome the limitations of conventional crossing points segmentation method, Hough transform is proposed to detect the skew angle of the warp and weft, and then coordinate transform is applied for the pixels in the fabric. By accumulating the gray levels of the pixels in the same horizontal or vertical line in the fabric image after coordinate transform, the projection curve can be obtained. By locating the minimums in the projection curve, the warps and wefts can be segmented individually. The crossing points can be located by combining the warp and weft segmentation results. The experimental results show that the algorithm proposed in this study can locate the cross points in fabrics with skew weft or warp yarns.