Enhanced beetle antennae search algorithm for spot color prediction

基于增强天牛须搜索算法的专色配方预测方法

¹ School of Water Resources and Hydropower, Xi'an University of Technology, Xi'an 710048, China
² School of Civil Aviation, Northwestern Polytechnical University, Xi'an 710072, China
³ School of Mechanical and Precision Instrument Engineering, Xi'an University of Technology, Xi'an 710048, China
⁴ School of Communication Engineering, Shenzhen Polytechnic, Shenzhen 518000, China

Received: 31 March 2022

Abstract

专色的准确预测是包装印刷领域的重要技术之一。为了得到更加准确的专色配方, 提高专色配色精度, 提出了一种结合最小二乘法和增强天牛须搜索算法的专色配方预测方法, 并利用吸光度来解决专色配方的预测问题。研究了高透光特性PET薄膜的光谱模型, 并构建了吸收光谱机理模型; 提出了增强天牛须搜索算法, 在传统天牛须搜索算法的基础上, 引入突变概率项和方向修正项, 提升算法的搜索能力和收敛速度; 利用最小二乘法优化配色色域空间, 降低基色搜索维度, 提高寻优效率。应用所提出的增强天牛须搜索算法求解各基色比例, 预测专色配方, 并与传统天牛须算法、粒子群算法和蚁群算法进行比较, 验证所提方法在专色预测方面的有效性和优越性。研究结果表明, 所提方法与现有的3种方法相比, 具有更高的精度, 原有专色和预测专色之间色差均小于3, 且90%的色差小于1, 40%的色差小于0.1, 所提方法对于提高专色油墨的配色精度具有显著效果, 可准确地预测专色配方。

摘要

Color prediction is one of the most important techniques in the field of packaging and printing. A new prediction method combined least squares and enhanced beetle antennae search algorithm is proposed to obtain precise special color formula by using absorption spectrum. This paper focuses on the spectral model of high light transmittance PET films and constructs the absorption spectral mechanism model for color prediction. Secondly, an enhanced beetle antennae search algorithm with direction correction term and mutation probability term is proposed to improve the searching performance and increase the convergence rate. Thirdly, for promoting the search efficiency, the least squares method is used to optimize the color gamut space and reduce the dimension of primary colors. Finally, the enhanced beetle antennae search algorithm is applied to solve color formula and predict spot color. The effectiveness and superiority of the proposed method are validated in comparison with basic beetle antennae search, particle swarm optimization and ant colony optimization. The results illustrate that the proposed method is superior to the compared methods. The color differences between the original spot color and the predicted spot color are less than 3, in which 90% color differences results are less than 1, 40% color differences results are less than 0.1. All the results confirm that the proposed method has significant effect on spot color prediction and can predict the special color formula accurately.