基于光谱的多色分色中颜色分区判断算法的研究

Color Subarea Judgment Algorithm in Multi-color Separation Based on Spectrum

目的研究基于光谱多色分色中颜色分区的判断算法。方法提出一种基于光谱的颜色分区判断算法,该算法引入特征光谱的提取和匹配算法,并在此基础上设计颜色分区判断算法。针对该算法流程中的参数,设计实验进行推算。结果将光谱曲线的值域范围划分为12个等级,目标光谱与分区的距离和目标光谱与所有分区距离最小值的差值小于14%,预测光谱与目标光谱的RRMS小于0.01,色差小于1.5时,能够得到最佳的分区判断结果,每个色块的判断次数为1.93,约为遍历算法的24%。结论提出的基于光谱颜色分区算法能够有效地对目标光谱所在分区进行准确判断,简化基于光谱的多色分色算法流程。

The aim of this study was to investigate the color subarea judgment algorithm in multi-color separation based on spectrum. A color subarea judgment algorithm based on spectrum was proposed. The color subarea judgment algorithm was designed by introducing characteristic spectrum extracting and matching. On account of the parameters in the algorithm progress, the experiments were designed for calculation. The optimal subarea judgment results could be obtained under the following conditions： the value range of characteristic spectrum was divided into 12 levels, the differences between the target spectrum from subarea and target spectrum from the minimum distance of all subareas were smaller than14%, while the RRMS and color difference between the predicted and target spectra were smaller than 0.01 and 1.5,separately. The exam indicated that the judge count of each patch was 1.93, which was about 24% volume of the circulation algorithm. The proposed color subarea judgment algorithm based on spectrum could exactly judge the target subarea where the color existed in an efficient manner, which simplified the multi-color separation progress based on spectrum.