应用Kubelka-Munk双常数理论的数码转杯纱混色效果预测

为研究数码转杯纺的混色效果，利用红、黄、蓝三原色粗纱纺制混色纱并织成织物，测量其颜色特征值。在此基础上利用最小二乘法及相对值法分别求出单色纤维的吸收系数及散射系数，建立各自的Kubelka-Munk双常数混色模型。采用这二种方法建立的模型对混色织物的颜色及有色纤维的混合比例进行预测。结果表明：利用最小二乘法建立的模型对样本预测的平均色差为0.896，平均比例误差为2.84%；利用相对值法建立的模型对样本预测的平均色差为1.35，平均比例误差为3.04%；与相对值法相比，利用最小二乘法建立的Kubelka-Munk双常数混色模型可更好地预测数码转杯纱混色效果及混色纤维的比例。

Prediction of color blending effect of digital rotor yarn based on Kubelka-Munk double constant theory

In order to study the color blending effect of digital rotor spun yarn, color blended yarn were spun by red, yellow and blue rovings, and woven into fabrics, and the color characteristics of the fabrics were measured. On this basis the absorption coefficient and the scattering coefficient of the monochromatic fiber were calculated by the least squares method and the relative value method respectively, and two types of Kubelka-Munk double constant color mixing model were built according to the above mentioned methods. The fabrics color and blending ratio of colored fibers were predicted by those two models. The results show that the average color difference of the model established by the least squares method is 0.896 and the average ratio error is 2.84%. The average color difference of the model established by the relative value method is 1.35, and the average ratio error is 3.04%. Compared with the relative value method, the Kubelka-Munk double constant mixing model established by the least squares method can better predict the color blending effect and mixing ratio of digital rotor spun yarn.