Developing an image analysis based method for determining the number of filaments in synthetic multifilament yarns

The developing qualities of the products due to the rapidly changing needs and demands in proportion to the increasing population over time cause textile consumption to increase. Natural fibers can no longer meet the required demand. In order to close the unmet demand gap, chemically synthetic fibers have been obtained that increase the utilization rate day by day. Filament yarns produced by various chemical methods are formed in a continuous structure by passing through the straights in a continuous state without cuts. When producing multifilament yarns, variations are observed in the number of filaments of the yarns produced due to the blockage of the nozzles during production or the fiber breakages caused by tension. Since multifilament yarns also affect the yarn and consequently the fabric structure at a significant level, the filament number determination test has great importance among the standard physical tests. Today, the number of filament determination test is carried out manually in laboratories by counting visually without the use of a device depending on the operator. This situation is open to errors caused by the operator and also causes a loss of time. Within the scope of the study, a method based on image analysis is developed for the determination of the number of filaments and it is aimed to prevent operator-based errors to make it suitable for automation and to save time. In the study, firstly, laser beam intersection was made by separating the fibers from each other by static electricity. Then, the filaments that separated and intersected with the laser beam were obtained with the help of a camera. The images taken were counted with the help of an image analysis-based program. The validation of the new method was carried out by the method of cross-sectioning of multifilament yarns. Results showed that the newly developed method has a great advantage over the conventional one.