Premium
A harmony‐based approach to generating sequential color schemes for maps
Author(s) -
Chen Taisheng,
Zhu AXing,
Wu Mingguang,
Chen Menglin,
Zhang Mingda,
Jiang Weixing,
Lu Yue,
Wang Hong
Publication year - 2020
Publication title -
color research and application
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.393
H-Index - 62
eISSN - 1520-6378
pISSN - 0361-2317
DOI - 10.1002/col.22453
Subject(s) - artificial intelligence , computer science , color difference , harmony (color) , color space , pattern recognition (psychology) , computer vision , mathematics , image (mathematics) , art , enhanced data rates for gsm evolution , visual arts
Generating high‐quality sequential color schemes for maps is challenging for most mapmakers. This article presents a harmony‐based approach for automatically generating sequential color schemes for maps. The proposed approach consists of two steps: (a) color harmony quantification for sequential color schemes on maps and (b) color scheme generation using quantified color harmony factors. The approach was tested using three experimental maps, including the Chinese water body density index map, Chinese vegetation cover index map, and Chinese population density map. Effectiveness was evaluated by comparing color schemes generated by the proposed approach with those created by the probability‐based method and ColorBrewer. Twenty‐six participants were invited to rate each map color scheme using a 5‐point scale. A t test was also used to examine the significance of the difference. The results show that the mean points of color schemes using the proposed approach are higher than those using the probability‐based method, and the corresponding P ‐values are far lower than .05, which suggests that the proposed approach is better than the probability‐based method and can improve the sequential color scheme quality in automatic ways. The mean points of color schemes created using the proposed approach are also slightly higher than or similar to those of ColorBrewer, which are well‐known map color schemes manually designed by an experienced cartographer. This result further suggests that the proposed approach can meet the requirement of automatic generation for high‐quality sequential color schemes on maps.