Identification and formalization of knowledge for coloring qualitative geospatial data

Abstract

Creating a satisfying qualitative color scheme from scratch may be difficult for novice mapmakers and experts. A probability-based method is proposed to identify knowledge regarding qualitative color selection from readily available color schemes and formalize the discovered knowledge to assist in color creation. An unsupervised method to extract the general trends of color selection is presented, and the issue of qualitative color selection is translated into a multi-constraint optimization problem. A feasible solution for achieving the global optimum is then provided. A probabilitybased method is also proposed to match abstract color values with specific mapping features. This proposed approach is evaluated in a case study. The results of the case study suggest that the proposed method allows users to create qualitative color schemes more efficiently and confidently. K E Y W O R D Sinteraction, kernel density estimation, map color design, multi-constraint optimization, qualitative color schema wileyonlinelibrary.com/journal/col

A harmony‐based approach to generating sequential color schemes for maps

“…However, these rules and cartographer experience are difficult to directly quantify. To address this problem, we adopt a probability‐based color selection method developed by Wu et al This method was designed to estimate the probability of colors for maps based on color samples using kernel density estimation (KDE). The probability of a color (c) can be expressed as where x i is the color in the samples, and the number of color samples is n ; h is the bandwidth, which is the smoothing parameter; and K is the kernel function.…”

“…However, these rules and cartographer experience are difficult to directly quantify. To address this problem, we adopt a probabilitybased color selection method developed by Wu et al 11,15 This method was designed to estimate the probability of colors for maps based on color samples using kernel density estimation (KDE). The probability of a color (c) can be expressed as…”

“…Good sequential color schemes can increase usability, improve the map aesthetics, and meet the goals of clearly expressing ordered relationships. [9][10][11] These schemes, however, are difficult to create for particular ordered datasets. [12][13][14][15] For example, spectral color schemes that lack perceptual ordering are frequently misused to express ordered data, and map users are easily confused by them.…”

“…For instance, progressive colors from red to yellow are often adopted to represent the population density on thematic maps. Good sequential color schemes can increase usability, improve the map aesthetics, and meet the goals of clearly expressing ordered relationships . These schemes, however, are difficult to create for particular ordered datasets .…”

An adaptive approach to create on‐demand color schemes for mapping quantitative geographic data

“…Mellado et al introduced learning samples to guide color selection. Wu et al introduced a probability‐based method that was also based on identifying samples in which any colors could be used to visualize a concrete mapping feature but with varying probabilities. Wu et al suggested using the kernel density estimation (KDE) proposed by Silverman in 1986 to estimate these probabilities.…”