论文总字数：17443字

摘 要

随着社会快速进步，人们对于织物的品质、印染效果的要求也逐步提高，已经从少样式大生产转变为多样式小生产。印染厂对于市场趋势的变化反应速度也在逐步加快，这就要求印染配方的制定更快、更准、生产成本更低。而优化计算机配色算法是解决以上问题方法之一。计算机配色囊括了物理学、生物学、心理学等多种学科的综合技术，随着近90年测配色理论和计算机技术的快速发展，测色配色技术已经有了长足进步，计算机配色已成为信息化纺织加工、智能化纺织加工的重要一环。计算机测色配色带来的好处有很多：有效降低小样试染次数、通过颜色型号的标准化提高客户与印染厂沟通色样时的效率、染色准确率高、印染数据库实时更新、配色越准确率随数据库迭代程度发展而提高等等。计算机染色配色不仅降低生产耗能、提高了生产效率，更提升印染厂的竞争力。

本文通过大量论文查找、阅读、数据分析后发现，单色光的三刺激值与浓度的测定与混色后的三刺激值的数值关系不是理想中的线性加和，如果计算时依然使用线性加和便会带来染料配方浓度误差较大、染色后色差较大，不满足实际生产的要求。为探索AI技术在染料配方的实际应用，本文通过实验传统配色流程，尝试解决K/S值随浓度变化非线性对于染料配方带来的影响，分析了K-M模型的具体计算方法、原理和公式推导。结果表明，K/S随浓度的非线性表达会给染料配方带来误差。经过以上探索发现，目前AI人工智能应用场景依赖于海量数据，对于染色配方数据库的数据整合有很好的应用场景，但还不足以支撑构建数学建模等进一步应用。

关键词：三刺激值，计算机配色，K-M模型，非线性加和

Printing and Dyeing Color Formula Fata AI Exploration

Abstract

With the rapid progress of the society, people's requirements for the quality and effect of printing and dyeing of fabrics have gradually increased, and the large-scale production of less patterns has been transformed into the large-scale production of diversified patterns.Printing and dyeing factories to the changes in the market trend response speed is also gradually accelerated, which requires the formulation of printing and dyeing formula faster, more accurate, lower production costs.Optimizing computer color matching algorithm is one of the methods to solve the above problems.Computer color matching is a comprehensive technology that includes physics, biology, psychology and other disciplines. With the rapid development of color matching theory and computer technology in the past 90 years, color matching technology has made great progress. Computer color matching has become an important part of informationized textile processing and intelligent textile processing.There are many benefits brought by computer color measurement and color matching: effectively reducing the number of sample test dyeing, improving the efficiency of communication between the customer and the dyeing factory through the standardization of color space, high dyeing accuracy, real-time update of printing and dyeing database, and increasing the accuracy of color matching with the development of database iteration.Computer dyeing and color matching not only reduces production energy consumption and improves production efficiency, but also improves the competitiveness of printing and dyeing factories.In this paper, through a large number of papers found after searching, reading, data analysis, the monochromatic light tristimulus values and the determination of concentration and mixed color tristimulus values after the numerical relationship is not the ideal linear and, if the calculation is still using linear and will bring error formula dye concentration, dyeing color difference is bigger, can not meet the needs of practical production requirements.

In order to explore the practical application of AI technology in dye formula, this paper tried to solve the influence of the nonlinear change of K/S value with concentration on the dye formula through the traditional color matching process, and analyzed the specific calculation method, principle and formula derivation of the k-m model.The experimental results show that the nonlinear expression of K/S with the concentration will bring errors to the dye formulation.After the above exploration, it is found that at present, the application scenarios of AI artificial intelligence depend on the sea volume data, and there are good application scenarios for the data integration of dyeing formula database, but they are not enough to support the further application such as the construction of mathematical modeling.

Keywords: Tristimulus, computer color matching, K-M model, nonlinear summation

目录

摘 要 Ⅰ

Abstract Ⅱ

第一章 文献综述 1

1.1引言 1

1.2国内外研究背景 1

1.2.1国内研究背景 1

1.2.2国外研究背景 1

第二章 光学模型分析 3

2.1光度学原理 3

2.2色度学原理 3

2.3配色方法 4

第三章 计算机配色基础理论 6

3.1颜色的基本特征 6

3.2计算机测色方法 6

3.3视觉感知感原理 7

3.4颜色的混合 8

3.5视觉感知心理学 9

第四章 计算机配色的算法分析 10

4.1Kubelka-Munk光学函数 10

4.2匹配三刺激值 12

4.3色彩空间 13

4.3.1 XYZ色彩空间 13

第五章 实验部分 14

5.1实验原理 14

5.2实验原料 15

5.3实验步骤 15

5.4实验结果与讨论 15

5.5实验结论 16

参考文献 18

致谢 20

第一章 文献综述

1.1引言

计算机测色配色系统是集合了测色仪、计算机处理器、配色软件、数据库四个子系统于一体的系统^[1]，通过人机对话可以用于配方预测。自从1930年国际照明委员会提出色彩空间标准^[2]，理论研究、计算机的计算能力和软件交互就在逐步满足染色配色的实际生产需求。由于计算机的引入，也使得织物的染色配色效率和创新性有很大的提升，如可以小批量快速地生产、在较短时间找到性价比最高的配色处方、根据现有染料配色等等，且能够更快地响应市场变化，让染色纺织厂做到科学运营，根据服装品类和用户消费习惯等下游因素决定上游生产。销售数据化、可视化；通过大数据统计分析得到服装流行趋势，颜色流行趋势，做到预测超前；在提高投入产出比的前提下提高自动化水平，逐步实现去人工化。

请支付后下载全文，论文总字数：17443字