Make sure that you have all dependencies installed (see below).
To start the GUI, run:
python colormap_selector.pyRight-click into the two cross section plots on the left to change the start/end color of the colormap.
Use the sliders to select a different value of
L(= lightness) for each cross section.To quit, press the
Esckey or close the main GUI window.
Whenever the colormap changes the 3D view on the right shows a linear interpolation of the start/end colors in CIELab color space. Moreover, the sample scatterplot (displayed in the separate window) is adjusted to use the new colormap.
Currently the following dependencies are needed.
- Vispy (recent development version > 0.3.0)
- Matplotlib
- PyQt
This is a GUI to facilitate the design of colormaps with "good" properties. It originated from a discussion discussion on the matplotlib mailing list about designing a new default colormap.
One of the major problems with many existing colormaps (including the infamous and harmful rainbow/jet colormaps), is that they are not perceptionally linear. This means that equal changes in the values of the displayed data do not correspond to color changes that are perceived as equal by humans, which is particularly harmful for scientific applications. For details see the references below.
The most commonly used RGB color space is not well suited
for colormap design. Instead, there are other color spaces which are
specifically designed to be perceptually linear. One example is the
CIELab color space
which represents each color using three the parameters L (=
lightness), a (= red/green component) and b (= yellow/blue
component).
This GUI helps defining perceptually linear colormaps. It consists of
one main window which on the left displays two cross sections through
CIELab space. Each of these corresponds to all those colors with a
fixed value of L (lightness) that are representable in RGB
space. The L-value for each cross section can be changed using the
slider underneath.
The start/end color of the colormap can be selected by right-clicking inside the cross sections. The two selected colors are then interpolated linearly and this defines the whole colormap. The 3D view on the right illustrates this linear interpolation.
In addition, a sample scatterplot is displayed in a separate window which is adjusted whenever the colormap changes.
A random (and highly incomplete) selection of articles on color, colormaps and scientific visualization:
- Subtleties of color (Part 1 of 6) http://earthobservatory.nasa.gov/blogs/elegantfigures/2013/08/05/subtleties-of-color-part-1-of-6/
- Matplotlib: choosing colormaps http://matplotlib.org/users/colormaps.html
- Mycarta ("A blog about Geophysics, Visualization, Image Processing, and Planetary Science"), Category: Visualization http://mycarta.wordpress.com/category/visualization/
- Rainbow Colormaps – What are they good for? Absolutely nothing! http://medvis.org/2012/08/21/rainbow-colormaps-what-are-they-good-for-absolutely-nothing/
- Why Should Engineers and Scientists Be Worried About Color? http://www.research.ibm.com/people/l/lloydt/color/color.HTM
The currently selected color in each of the 2D plots on the left should be indicated by a white dot. Sometimes these color indicators are not visible because the dot is drawn underneath the cross section. This is due to a known bug in Vispy (see: https://groups.google.com/forum/#!topic/vispy/z8b-Wglo3j0)