Skew detection and yarns density calculation for woven fabric

Automatic identification of fabric structure is a vital research according to the fabric texture. Due to the skewed phenomenon which is inevitable during the scanning process, Hough transform is utilized for skew detection. Then wavelet filter is proposed to separate warps from wefts to enhance the information in vertical and horizontal direction, respectively. Finally, the gray projection curve including peaks and valleys is obtained in warp and weft directions. According to the peaks, the yarns can be located and segmented apparently and the fabric density could be obtained. Experimental results show that the precision of skew detection could be controlled within 2° while the accuracy of yarns density detection can reach up to 100%, which demonstrate that the proposed method is effective in skew detection and fabric density calculation.