Abstract:

Objective Market demand for textiles is ever-increasingly diversified, and the performance of yarns, as an early product in textile production, impact final product quality. It is feasible to simulate the performance and structure of yarns under different materials and processes, which can be used to guide the flexible production of textiles. This endeavor is crucial as it helps optimize the production process, reduce costs, and improve the quality and consistency of textile products, thereby enhancing the competitiveness of the textile industry.

Method Based on the imaging principle of two-planar mirrors, this research proposed a three-dimensional simulation method for yarn cores. The yarn's real image and four virtual images from mirror reflection were captured as multi-perspective images, which represent the five views of yarn. After applying the Otsu thresholding, morphological opening and dilation methods, the smooth yarn core binary images were acquired. Next, five circles were drawn with the widths of yarn core in five perspective images as diameters, respectively. Then, according to the geometric principle of mirror reflection, the circles obtained from each perspective were moved to fit the yarn core cross-section.

Results To verify the effectiveness and accuracy of the proposed method, the cross-sectional area and the coefficient of variation of the reconstructed yarn model were calculated. Given the widespread use of the USTER^® TESTER 5 in yarn appearance evaluation, we decided to compare the measured values obtained from our method with its measurement results. This comparison serves as a means to evaluate the accuracy of the proposed method’s modeling at the two-dimensional. In addition, the measured values were also compared with three-dimensional measurement method to evaluate the effectiveness of the proposed method at the three-dimensional level. At the two-dimensional level comparison, the measured values of the proposed method were compared with the USTER^®TESTER 5 measuement results. The cross-sectional area of the reconstructed three-dimensional model of yarn was measured. The results showed that the correlation coefficient between the average cross-sectional area of the reconstructed yarn model and the diameter measured by the USTER^®TESTER 5 was as high as 0.996, and the correlation coefficient between the coefficient of variation of cross-sectional area and the coefficient of variation of the uniformity of the yarn measured by the USTER^®TESTER 5 was 0.834. At the three-dimensional level comparison, the results were compared with the three-dimensional measurement method instead. The correlation coefficient between the measured values of the proposed method and the three-dimensional measurement method was 0.965, which indicates the positive correlation of the measured values. In addition, the uniformity of short segments can be observed from the change in area. By using the proposed method, the task of simulating the three-dimensional model of yarn cpuld be fulfilled, and the synthesized three-dimensional model was close to the irregular shape of a cylinder, which could effectively reflect the unevenness of yarn. This fully demonstrated the feasibility of the reconstruction method for the three-dimensional model of yarn core.

Conclusion This paper presents a method for constructing a three-dimensional model of yarn based on the assumption of irregular circular cross-section of yarn. The obtained results demonstrate a high correlation with the USTER^®TESTER 5 at the two-dimensional level and a positive correlation with another three-dimensional method at the three-dimensional level, thereby clearly indicating the effectiveness of the proposed method. In addition, the simulated yarn model enables the observation of structural characteristics from different angles, allowing for acquisition of more detailed information and thus presenting excellent application prospects. However, a major drawback of this method is relatively slow for reconstruction, which needs to be furthrt improved. Such a method has significant implications for the textile industry, as it provides a more accurate and detailed way to analyze and understand the properties of yarns, which can ultimately lead to improved product quality and performance.

Key words: yarn core, yarn evenness, image processing, three-dimensional simulation, three-dimensional measurement