基于结构相似的CIE L*a*b*颜色信号向孟塞尔色序系统分级转换

针对传统数学建模法无法实现CIELAB颜色信号向孟塞尔色序系统精确转换的问题,提出一种基于结构相似的分级转换算法。首先使用Pearson相关系数分析CIELAB颜色系统与孟塞尔色序系统间的结构关系,接着在此基础上采用分级转换的方式先后得到CIELAB颜色信号对应的孟塞尔明度V、色调H和彩度C三色标,并在第二级转换求色标H时使用了一种多次插值求最小色差的新方法,以保证算法的计算精度。对2735组检验样本的测试结果表明,本文算法具有良好的计算精度和可靠性,并可被广泛应用于各类图像颜色信号处理系统中。

Hierarchical conversion algorithm of color signal from CIELAB space to Munsell system which based on structure correlation analysis

In order to solve the inaccuracy problem of traditional color-signal conversion methods from CIELAB space to Munsell system, a new hierarchical conversion algorithm based on structure correlation analysis was proposed. Firstly, this article evaluates the structure correlation between CIELAB space and Munsell system by using the Pearson coefficient. Then on the basis of structural correlation evaluation, this article uses a hierarchical conversion algorithm to get the corresponding Munsell V, H, and C color mark of the CIELAB color signal. In addition, the algorithm proposes a new method by choosing the nearest interpolation H values during the second conversion step to ensure the calculation precision. The experimental results of 2735 test samples show that the new algorithm has good calculation accuracy and reliability, and can be widely used in different kinds of color signal transmission and image processing system.