基于单常数Kubelka-Munk理论的棉纤维颜色预测

Cotton fiber color prediction based on the single constant Kubelka-Munk theory

为解决混色棉纤维颜色预测效果较差、模型预测精度较低的问题,文章通过对Kubelka-Munk理论的分析,确定采用基于单常数Kubelka-Munk理论的配色模型对混色棉纤维进行预测。在明确此配色模型不足之处的基础上,提出了标准化映射的方法对模型进行适用性改进,消除极端K/S值对模型预测效果产生的不利影响,从而改善模型的颜色预测效果。通过试验发现,模型对混色棉纤维颜色预测的适用性有明显改善,改进后的单常数K-M模型对混色棉纤维颜色预测精度有大幅度提高。

In order to solve the problems of poor color prediction effect of mixed-color cotton fiber and the low accuracy of model prediction,it was determined that the color matching model based on the single constant Kubelka-Munk theory was used to predict the mixed-color cotton fiber.On the basis of clarifying the shortcomings of this model,the method of standardized mapping was put forward in this paper to improve the applicability of the model,eliminate the adverse effects of extreme K/S values on the prediction effect of the model,and thus improve the color prediction effect of the model.Through the experiment,it was found that the applicability of the model to the color prediction of mixed-color cotton fiber was significantly improved,and the improved single constant K-M model greatly improved the color prediction accuracy of mixed-color cotton fiber.