Detection And Application Of Yarn Hairiness By Digital Self-rotating Image Method

Yarn hairiness is one of the important indicators for yarn quality evaluation and testing.Reasonably controlling hairiness during textile production can effectively avoid the adverse effects caused by hairiness,and it has great significance for improving the quality of textiles.In order to further improve the accuracy and practicability of hairiness detection methods,this paper designs and develops a yarn self-rotating image acquisition system suitable for yarn image processing in view of the poor repeatability of the test results of existing yarn hairiness detection methods.First,a device for acquiring yarn images is proposed.According to the calculation of real yarn hairiness parameters,images of multiple angles of the yarn need to be processed,and a yarn self-rotating image acquisition system is designed;detecting the appearance parameters of the yarn according to the image method requires the outline of the yarn image to be clear,and the light source of the image acquisition system.Adopt the back-irradiation method.The proposed image acquisition device can be divided into a yarn transmission structure and a yarn image imaging structure.The device can acquire high-quality images of yarn image sequences of different angles.Secondly,the collected image of yarn is processed by computer image processing algorithm.The background difference processing method is used in the yarn background processing,and subsequent image processing is performed based on the background difference processing.Yarn image tilt correction improves the accuracy of subsequent image calculations;the use of Wiener filtering to process yarn images makes the contrast between the yarn and the background more pronounced;for the direct threshold segmentation processing image,it is difficult to accurately segment the yarn image,an adaptive gray The degree adjustment algorithm combined with the maximum inter-class variance threshold segmentation method achieves accurate segmentation of the yarn image;a differential morphological processing algorithm is proposed during the extraction of the yarn trunk and yarn hairiness,which can accurately separate the yarn trunk and yarn hairiness It is extracted to lay a good foundation for the subsequent calculation of yarn parameters.Then,calculate the actual length and quantity of yarn hairiness.The calculation method of the true length and quantity of yarn hairiness is deduced according to the variation rule of the projected length when the object rotates.Using this method,the true length and quantity of yarn hairiness can be obtained.The proposed detection method can detect yarn indexes such as the yarn trunk CV value while detecting the length and quantity of yarn hairiness.Finally,the results of the proposed image method are compared with those of the traditional yarn hairiness detection method.The experimental results show that the results of the proposed method have a strong correlation with the results of the traditional method.It shows that the image method can be used to detect the appearance parameters of the yarn and can be used as a guide for industrial production.Multi-angle image method for testing yarn hairiness parameters can provide convenience for studying yarn hairiness morphology and subsequent processing of yarn.The proposed image acquisition device and image processing algorithm can provide a theoretical basis for the application of machine vision in yarn appearance parameter detection.