Skew Correction and Density Detection of Knitted and Woven Fabric

Automatic identification of fabric structure is a vital area of research. The skewing phenomenon is inevitable during the scanning process, so the fabric skew correction method based on projection profile analysis is proposed. First, Butterworth low-pass filter is applied to remove noises after skew correction of the fabric image. Then, power spectrum is obtained by Fast Fourier Transform (FFT), in which the peaks are extracted from the vertical and horizontal direction, respectively. Finally, the reconstructed image is obtained via Inverse Fast Fourier Transform (IFFT) according to the peaks, so that the information of warp and weft can be separated to calculate the warp and weft density. Experimental results show that the accuracy of the skew correction can be controlled within [−1◦, 1◦], and the detection accuracy of yarn density can reach 98%. The proposed method can accurately detect skew angle and density of woven and knitted fabrics.