A visual distortion detection method for textile cloth

At present, most textile enterprises in China have a high degree of labor intensity, and the defect detection of textile still depends on manual inspection. This way has many disadvantages such as strong subjectivity and high labor intensity. With the growing maturity of machine vision technology, it has been widely applied to the industrial production process. The paper deals with the problem of twisting and deformation of textile cloth driven by cloth winding motor. Firstly, CCD industrial camera was used to capture the image. Secondly, a series of operations was applied to the image to extract the features of the warp and weft of cloth, such as the binarization, morphological open and close, edge detection and Hough transform. Finally, the weft inclination angle was calculated to help detect the distortion of the cloth. Experimental results showed that, the inclination angle detected by the proposed algorithm is consistent with that measured by protractor, and the maximum error is −3.57%.