Research On The Color Prediction Method Of Colored Cotton Fiber

In the era of rapid development of science and technology,higher requirements for work efficiency are proposed.In order to overcome the inefficiency of manual work in different jobs,enterprises are also more and more inclined to choose a method that combines computer technology and machines to improve production efficiency and reduce labor costs,as is the case in the field of color spinning.Traditional artificial color matching has not been able to meet the color matching efficiency requirements of color spinning companies.The use of computer color matching technology instead of artificial color matching technology,thereby improving color accuracy and color matching efficiency,which not only has become the best choice of color spinning enterprises,but also the inevitable trend of color spinning.However,the key that computer color matching technology cannot be widely used in color spinning enterprises nowadays is that the color matching models have the problem of low color prediction accuracy.Aiming at this problem,three different types of color prediction models are discussed in this paper,the applicability and prediction accuracy of different models are analyzed,and finally the best color prediction model for color prediction of mixed-color cotton fibers is determined.Firstly,about the Stearns-Noechel model,according to the different types of model parameters,the model is divided into numerical model and relational model.The prediction effects of two models with different parameter forms on mixed cotton fibers are compared,and the overall prediction effects of the two parameter models on mixed cotton fibers are relatively close.In the comparison of model applicability and prediction accuracy,it is found that the applicability of different parameter forms is different,and the prediction accuracy is also obviously different.Secondly,for the Friele model,it was verified through experiments that the Friele model had poor applicability in color prediction,and the applicability of the Friele model was optimized through the loop parameter method.Thirdly,the color prediction model based on Kubelka-Munk theory is discussed.The applicability of double constant Kubelka-Munk model to the color prediction of mixed cotton fibers and the accuracy of the model are analyzed.It is determined that multiple uncertainties in the model have greater influences in its color prediction.In the analysis and discussion of the single constant Kubelka-Munk model,the problem affecting the prediction accuracy of the model was found from the analysis of the prediction formula of the model,and a standardized mapping method was proposed to improve it.As a result,a better optimization effect was obtained.The three color prediction models are compared in terms of model applicability and model prediction accuracy.It is determined that the optimized single-constant Kubelka-Munk model is the color prediction model with the best comprehensive prediction effect.Based on the actual problems of the dyed spinning enterprise,the industrial application of the color prediction model is discussed.First,the visualization of the spectral reflectance is proposed to optimize the customer experience.Second,the colored cotton fibers that are different from the test sample dyeing process are used to verify the single constant.Finally,for the problem that the least square method cannot be directly used to solve the standard sample formulation,the reflectance difference method is proposed to locate the types of monochrome fibers,and the formula of standard sample from the customers can be completely fitted when it is used in combination with the least square method.