| The hybrid multifilament is a kind of chemical multifilament composed of monofilaments of different colors.The spinning process is based on the chip spinning method of stock solution coloring spinning method,which benefits from the spinnability of microfiber,and makes use of the three primary color mixing principle and the color mixing characteristics of human visual space.The monofilament with different colors can be spun with less color masterbatch.By adjusting the color,thickness and color mixing ratio of the monofilament in the multifilament,the hemp gray hybrid multifilament or the plain hybrid multifilament with uniform color can be obtained.With the global demand for colored filaments increasing year by year,the development prospect of this spinning process is very promising.However,the spinning process is still in the trial spinning stage,lacking theoretical guidance,computer-aided design and color matching system.Because the final color rendering effect of the hybrid multifilament is related to its surface color uniformity,in order to analyze the factors affecting the surface color uniformity of the hybrid multifilament fabric,it is necessary to put forward not only the objective index to describe its surface color uniformity,but also the color prediction model that can characterize the surface color of different fabrics,so as to realize the soft proofing before spinning multifilament,and provide basis for the design and production of hybrid multifilament products.The research of this article mainly includes the following aspects:First,a color prediction model of hybrid multifilament is constructed.The color prediction model can not only calculate the overall mixed color of the sample,but also calculate the color of each point on the sample surface.Therefore,before constructing the color prediction model,a geometric structure model which can reflect the position and arrangement of different color monofilaments in the sample is established.When constructing the geometric structure model of multifilament,firstly,the position and arrangement of monofilaments in the cross section of multifilament are assumed,and it is considered that monofilaments are arranged in parallel along the viewing direction in the cross section of multifilament into several rows,and each row is called a sub-band.In order to reflect the position change of different color monofilament along the length direction of the multifilament,the multifilament is regarded as composed of several multifilament segments.The position arrangement of monofilament is the same in different multifilament segments,but the color distribution of monofilament is the same in the same multifilament segment,but different in different multifilament segments.Thus,the calculation method of the position arrangement of the monofilament in the cross section of multifilament,the color assignment method of monofilament in different positions,and the position adjustment method of the monofilament of different colors are proposed.In order to establish a model that can calculate the color of each point on the surface of multifilament,from the perspective of monofilament,a method for calculating the mixed color of two adjacent monofilament segments is proposed based on geometric optics knowledge and simple radiation theory,that is,the film stacking color prediction model based on geometric optics and the k-m iterative color prediction model based on simple radiation theory.After obtaining the mixed color of two adjacent monofilament segments,the mixed color of multiple monofilament segments is calculated by iterative calculation method,that is,the mixed color of monofilaments in the sub-band of the cross-section of multifilament,and it is also the color of a point on the surface of multifilament.The color of all points on the multifilament surface is calculated in the same way.Finally,the tristimulus value mixing algorithm is used to calculate the mixed color of all points on the surface of multifilament,that is,the overall mixed color of multifilament.In order to verify the calculation effect of the two color prediction model,a batch of black-and-white monofilament color mixing hybrid multifilaments with different monofilament color mixing ratio and multifilament fineness were spun.The verification results of the calculation effect of the film stacking color prediction model show that the mean color difference of the calculation results of the model is 9.66 CMC(2:1),and the main error comes from the assumption that the monofilament cross section is square when the model is established.Then,the initial calculated value of the model is modified by multiple regression algorithm.The mean color difference of the calculated result of the modified model is 0.98 CMC(2:1).The verification results of the k-m iterative color prediction model show that the average color difference of the initial calculation results of the model is 1.54 CMC(2:1),which can basically meet the actual production demand without correction.Second,a color prediction model of hybrid multifilament fabric is proposed.Because the fabric is closer to the final use state of the hybrid multifilament,and the factors affecting the color uniformity of the fabric surface are more complex than those affecting the color uniformity of multifilament,so the research on the color prediction model of the hybrid multifilament fabric has more practical significance.When constructing the fabric geometric structure model,firstly,the calculation method of monofilament position arrangement in the cross section of flattened yarn is established based on the multifilament geometric structure model,and then the calculation method of yarn arrangement gap and the method of adjusting the yarn position at the weaving point according to the fabric structure type are proposed.The calculation method of single filament mixed color in the fabric color prediction model is the k-m iterative color prediction model preferred in the multi filament color prediction model.In order to evaluate the prediction effect of k-m iterative fabric color prediction model on the color of each point on the fabric surface,the histogram similarity evaluation index,texture correlation difference evaluation index and the comprehensive evaluation index SFSC(similarity of fabric surface color)based on these two evaluation indexes are designed.The value range of histogram similarity evaluation index is 0 to 1.The larger the value,the better the prediction effect.In order to verify the calculation effect of k-m iterative fabric color prediction model,a batch of fabrics with different texture types were woven with the hybrid multifilament mixed with black-and-white monofilament.The verification results show that in the color calculation results of each point on the fabric surface by k-m iterative fabric color prediction model,the mean value of gray histogram similarity evaluation index is 0.88,the mean value of texture correlation difference similarity evaluation index is 0.21,and the mean value of comprehensive similarity evaluation index SFSC is 0.33,the mean color difference of the overall mixed color calculation results of the fabric is 0.84 CMC(2:1),which is less than the mean color difference of 1.01 CMC(2:1)calculated by Kubelka-Munk model,1.28 CMC(2:1)calculated by Stearns-Noechel model and 1.19 CMC(2:1)calculated by Friele model.By comparing and analyzing the prediction effect of k-m iterative fabric color prediction model on the color of each point on the fabric surface and the prediction effect on the overall mixed color,it is found that there is not a simple linear relationship between them,which shows that the overall mixed color prediction effect can not only be used as an index to judge the calculation effect of the color prediction model.This conclusion has also become the basis for the subsequent color matching algorithm.By analyzing the effects of fabric texture type,multifilament fineness and black monofilament color mixing ratio on the prediction effect of the color prediction model,it is found that the length of floating yarn,monofilament specific surface area,monofilament absorption and scattering coefficient have a certain influence on the color difference and surface color similarity of the calculated results of the color prediction model.The influence of fabric geometric structure model construction parameters on the calculation effect of k-m iterative fabric color prediction model is studied.The results show that the fabric geometric structure area has a significant influence on the calculation results of k-m iterative fabric color prediction model,while the number of yarn segments,the shape of yarn cross section and the adjustment method of monofilament position in yarn have little influence on the calculation results of k-m iterative fabric color prediction model.Next,the evaluation index of surface color uniformity of hybrid multifilament fabric is designed.By analyzing the uneven distribution of fabric surface color,the evaluation index of color difference mean color uniformity of each point color relative to the overall mixed color is proposed according to the relationship between each point color on the fabric surface and the overall mixed color,the evaluation index of effective gray level width is designed according to the number of gray level types of each point color on the fabric surface which affects the color uniformity evaluated by human vision,the evaluation index of color complexity is given according to the number of color types on the fabric surface and the complexity of spatial distribution,the evaluation indexes of texture entropy and texture homogeneity are proposed according to the entropy and homogeneity of gray level co-occurrence matrix,and a comprehensive color uniformity evaluation index is established based on the calculation results of each color evaluation index finally.At the same time,referring to the color difference grading method,the color uniformity grading method of the calculated values of each index is established.In this grading method,the lower the grade,the better the color uniformity.In order to verify the effectiveness of the evaluation index,the hybrid multifilament fabric with black-and-white monofilament mixed color is used.The observed value of fabric surface color uniformity is obtained through subjective evaluation experiment.The relationship between the observed value and the calculated value of each evaluation index is analyzed.The results show that the determination coefficients R~2 between the observed value and the calculated values of color difference mean evaluation index,effective gray level width evaluation index,color complexity evaluation index,texture entropy evaluation index and texture homogeneity evaluation index are 0.76,0.83,0.79,0.80,0.73 and 0.78 respectively.The results show that the calculation results of each index can well characterize the color uniformity of fabric surface,which is consistent with the observation results of human vision.Finally,the influence of the structure parameters of the hybrid multifilament fabric with black-and-white monofilament color mixing on its surface color uniformity is analyzed.It is concluded that the fabric structure type and warp and weft yarn arrangement density have no significant influence on the fabric surface color uniformity,while the multifilament fineness,monofilament number and monofilament color mixing ratio have significant influence on the fabric surface color uniformity.Last,the hybrid multifilament aided design and color matching system are preliminarily developed.Using k-m iterative color prediction model and fabric geometric structure model,the hybrid multifilament aided design system is preliminarily developed.The system can design the relevant parameters of monofilament,multifilament,yarn and fabric,and output and display the preview of the designed fabric,its overall mixed color chromaticity value and surface color uniformity.According to the relationship between the color difference and surface color similarity of the overall mixed color of simulated fabric and real fabric calculated by k-m iterative color prediction model,a double threshold cyclic color matching algorithm is proposed,that is,different color matching schemes are obtained by assigning fabric structure parameters,and then judge whether the color scheme meets the requirements of color difference threshold and surface color similarity threshold,and screen the color scheme.The color matching scheme not only includes the color mixing ratio of different color monofilaments,but also contains other parameters related to multifilament,yarn and fabric,which is more conducive to guide production practice.Generally,there is more than one color scheme that meets the requirements,so it is convenient for manufacturers to choose according to their actual production capacity.The effectiveness and application value of the preliminary developed aided design and color matching system are verified by example application and product design.The research conclusions of this paper can not only provide theoretical support for the research and development of hybrid multifilament products,but also provide ideas for the construction of color spinning color prediction model and fabric geometric structure model,and provide reference for the evaluation of fabric surface color uniformity. |
PDF Full Text Request