CONSELF | Post-processing color maps: pros and cons
8996
single,single-post,postid-8996,single-format-standard,ajax_fade,page_not_loaded,,columns-4,qode-child-theme-ver-1.0.0,qode-theme-ver-10.1,wpb-js-composer js-comp-ver-5.0.1,vc_responsive
 

Post-processing color maps: pros and cons

brewer

Post-processing color maps: pros and cons

Which is the best color map? This is a common question in the field of scientific simulation. But, even if it sounds really simple, it has not a universally accepted right answer. Some choices are popular among the community, but they are driven by tradition instead of effectiveness. And, as always, it is a trade off that depends on the context too. In this post we present a review of most common color maps based on some technical articles, listing pros and cons for each of them. We used as benchmark two standard CFD and FEM simulations: a car external aerodynamics and an axially loaded connecting rod. We also show how to change the color map inside CONSELF online post processor or in your local Paraview.

Rainbow

It is the most common color map. It has been the first choice among the scientific community and since then it became the standard one for visualization. Nevertheless, many scientific articles have been published where it is defined as “dangerous”, “misleading” or even “harmful”.

rainbow

Rainbow

Pros:

  • It is perceived as the natural one and provides results a “good looking” style. In practical cases, this can be really important, we all know how fundamental is to present results in the most entertaining way possible to catch our audience attention.
  • It allows to clearly identify distinct regions, thanks to the rainbow-based five colors progression: blue, cyan, green, yellow and red. It could even be increased up to six if violet is added. And despite the fact that boundaries between regions (colors) are arbitrary and uneven, it makes simple to compare different portions of the domain.

Cons:

  • The ordering is not intuitive, it is based on the wave length of visible light. Studies in the literature show that when human subjects are asked to order these colors, the results vary.
  • There is no direct correspondence between increments in the color space (RGB values) and perceptual increments (how different colors are perceived). This results into uneven variations of colors widths and steep transitions between colors, causing not equally distributed bands in the color map. When colors are used to represent variables, these shortcomings can be actively misleading, making the user perceive sharp transitions where there is none as well hide transitions that happen to fall within a band perceptually equivalent color.
  • The most intuitive parameter to guide color ordering is luminance, and the rainbow map is neither isoluminant nor ordered luminance. It goes from low luminance (blue) to high (cyan) to moderate (green) to high (yellow) to moderate (red).
grayscale

Grayscale

Grayscale

As already stated for the rainbow map, changes in luminance are easily perceived by the visual system. For this reason the grayscale map, simply going from black to white, is often found in the literature. It is standard for many medical applications, inherited from classical x-ray imagery.

Pros:

  • Color ordering is perceived very clearly.
  • High spatial contrast sensitivity.

Cons:

  • Bad absolute resolution. Luminance of a location is interpreted by the visual system relatively to its surroundings. Thus, it is difficult to get the absolute value represented by the color map. Studies in the literature show that users asked to guess an absolute value of a given variable based on surrounding reference, produce errors of up to 20%.
  • Inappropriate for 3D surfaces. 3D shading uses changes in luminance to render the third dimension in your screens. Thus, using grayscale color map on 3D objects will interfere with it.

Black body radiation

A color map similar to the grayscale is the black-body radiation, it mimics the visible color of a physical object as it is heated: a dark red represents the lowest value, the color gets brighter as the value increases, moving from red to orange to yellow to white.

blackbodyradiation

Black body radiation

Pros:

  • About all of the same pros as the grayscale.
  • The colors help to perceive variables absolute value in the map.
  • Results appear “good-looking”, especially for some specific applications.

Cons:

  • Even if luminance clearly defines ordering, color progression is opposite with respect to the one of the rainbow color map, and being the latter still considered standard this can be misleading.
  • Black coloring is still present and, as for the grayscale, it interferes with shading, making it inappropriate for 3D surfaces. It can be partially solved using the subset from dim red through white, enabling some shape perception on the entire object.

Isoluminant Green-to-Red & CIELAB

Another important category it the so called “isoluminant” color map. Here we present two isoluminant color maps: the Green-to-Red and the  “CIELab” isoluminant.

redgreenisoluminant

Isoluminant Green-to-Red

isoluminantcielab

Isoluminant CIELAB

Pros:

  • Does not interfere with shading.
  • Relative comparisons are consistent.
  • Absolute values perception is relatively accurate.

Cons:

  • They do not look good nor appealing.
  • They tend to look too dark.
  • It is difficult to catch small changes in hue for the visual system, so these maps have a low effective resolution.

Brewer CIELAB1 & Cool-to-Warm

From what we have seen so far it seems that one should choose between the blackbody, for 2D datas, and isoluminant, if shape is what you care about. In this debate, many experts discussed color choices in detail, and one interesting result is that the so called “Colorbrewer” map looks like an excellent middle ground, probably more useful for most mappings onto 3D surfaces than an isoluminant map, with minimum geometry distortion.

The two maps presented here, Brewer CIELAB and Cool-to-Warm, are generated starting with these considerations. They are divergent maps, meaning they go from color to neutral to color again, separating data into two groups around the median value and maximizing the color range.

brewer

Brewer CIELAB1

cooltowarm

Brewer (Cool-to-Warm)

Pros:

  • Intuitive color ordering. It goes from cool (blue) to neutral (white) to warm (red).
  • It is similar to the rainbow color map.
  • It is aesthetically pleasing.
  • It does not interfere too much with shading.
  • It does not hide data in bands as much as the rainbow.

Cons:

  • It is difficult to be edited by the user due to a lots of control points (9 to 11).
  • It is not quite as sensitive to change as the grayscale map.
  • It interferes a little more with shading than the isoluminant maps.
  • It has less color separation than the rainbow.

More color maps

Following images show you some additional color maps you can find in CONSELF online post processor or in your local Paraview.

bluetoyellow

Blue-to-Yellow

blackbluewhite

Black-Blue-White

rainbowdesaturated

Rainbow desaturated

blot

Blot

Change you color map in CONSELF and in Paraview

The video tutorial on the right shows you how to change the color map in CONSELF online post processor and in your local Paraview. Following these steps, you will be able to select the most appropriate to your needs very easily.

If you liked this article, share it and subscribe to the newsletter to stay tuned with our updates.

About CONSELF

CONSELF wants to make state of the art, cutting edge technology, available to every professional in the globe.

CFD & FEM simulation software is a very powerful tool, with its adoption optimization and innovation can be achieved in every field. To make this instrument accessible to everyone means lowering costs, but mainly to develop an infrastructure that favours a super-easy adoption by market new entrants.

CONSELF is highly committed and will pursue this goal working side by side with professional and industries to define the best strategies and solutions.

CONTACT:
Alessandro Palmas – alessandro.palmas@conself.com

No Comments

Post A Comment