Color Matching Research Based on Octree-LSD Method and Kansei Engineering: A Case Study of Typical Images of the Grain Rain

摘要

This paper proposes a method of extracting colors from natural images scientifically and then coordinating them to apply to corresponding situations. By applying this method, a case study of typical images of the Grain Rain, which is one of the 24 solar terms in Chinese traditional calendar was carried out. The method is as follows: looking for the typical images of the Grain Rain through literature review and social investigation, then taking pictures in the field. Using octree combined with least-significant difference method (octree-LSD method) to original colors, after that, several colors were picked of each image and the degree of beauty of them was also calculated according to the M·Spenser's aesthetic measurement. Only if it comes out that these colors will be qualified on the aspect of aesthetic when combining together, will the next step be carried out which is adopting Munsell Color Harmony Theory to determine the area ratio of each color and gain the color scheme. Last but not the least, using the Semantic Difference method in Kansei engineering to measure the color emotion of each color scheme, then color schemes which will be able to represent the Grain Rain and the application situation of them would be obtained. This method combines scientific calculation methods, western scientific color systems and empirical subjective opinions, which insure that the color schemes obtained by this method can not only be valid but also consistent with people's cognition.