Chapter 10: Color

Question 1

Through which channels do humans perceive color?

Answer 1

One from red to green, one from blue to yellow and one from black to white, encoding luminance information. The luminance channel conveys high resolution edge information while the red-green and blue-yellow channels are lower resolution.

Question 2

Why is the RGB color space not a good representation for how we see color?

Answer 2

The red, green and blue axes of the RGB color space are not useful as separable channels, they give rise to the integral perception of a color. They also do not encode colors like a human would do through the different color channels.

Question 3

What is the HSL system, and explain why it is, or is not, a good reflection of how we see color.

Answer 3

The hue-saturation-lightness system, is more intuitive and is heavily used by artists and designers. The hue acris captures what we normally think of as pure colors that are not mixed with white or black. The saturation axis is the amount of white mixed with that pure color. For instance, pink is a partially desaturated red. The lightness axis is the amount of black mixed with a color. Despite the popularity of the HSL space, it is only pseudo perceptual: it does not truly reflect how we perceive color. In particular, the lightness L is wildly different from how we perceive luminance.

Question 4

Name a color system that does actually reflect the way humans see color

Answer 4

color space that attempts to provide a perceptually uniform space is the lab* color space. L* (lightness) which ranges from 0 (black) to 100 (white) and corresponds to the human perception of lightness. a* (from green to red) and b* (from blue to yellow) which ranges from -127 to +127. These two components are used to represent the chromaticity of a color.

Another color space would be CIE 1931 XYZ. Another experimentation to represent all human visible color.

Question 5

To which colors / wavelengths are humans most sensitive?

Answer 5

Green (and yellowish). This is thought to be due to the fact that the human eye has more receptors for green light, called “cones,” in the retina than for other colors. Additionally, the peak sensitivity of these green receptors is around 555 nanometers, which is in the middle of the visible light spectrum, making it easier for the eye to detect.

Question 6

Explain Hue, and how it can be used as an encoding channel.

Answer 6

The HSL color system’s hue component refers to the dominant wavelength of light that makes up a color. It is used as an identity channel, specifically for categorical data and showing groupings. It has a high ranking among channels for representing categorical data, second only to spatial position. The number of distinguishable steps for hue in small, separated regions is moderate, generally around six or seven bins. Unlike saturation and lightness, hue does not have an inherent order, making it important to consider this when designing visualizations.

Question 7

Explain Saturation and how it can be used as an encoding channel.

Answer 7

The saturation component of the HSL color system can be used as a magnitude channel for ordered data. However, it has the problem of low accuracy when used for non-contiguous regions. The number of distinguishable steps for saturation is low, typically around three bins. Additionally, saturation interacts strongly with the size of the region being represented, making it more difficult to perceive in small regions than in large ones. Saturation and hue are not separable channels within small regions for the purpose of categorical color coding. To ensure distinguishability in small regions, designers should use bright and highly saturated colors. In contrast, when representing large regions such as backgrounds, it is recommended to use low-saturation colors.

Question 8

Explain luminance and how it can be used as an encoding channel.

Answer 8

The lightness component of the HSL color system, also known as luminance, can be used as a magnitude channel for ordered data types. However, using it as a magnitude channel has some limitations, such as low accuracy in perceiving whether non-contiguous regions have the same luminance due to contrast effects. This results in a small number of distinguishable steps, typically less than five when the background is not uniform. It is important to avoid using grayscale if more than two to four bins are required. Another limitation of using luminance as a magnitude channel is that once it is used, it cannot be used for other purposes. A crucial consideration when visual encoding with color is that luminance contrast is the only way to resolve fine detail and see crisp edges.

Question 9

What is the fourth channel that can be used for encoding?

Answer 9

The fourth channel strongly related to the other three color channels is transparency. Information can be encoded by decreasing the opacity of a mark from fully opaque to completely see-through. Transparency cannot be used independently to the other color channels because of its strong interaction effect with them: fully transparent marks cannot convey any information at all with the other three channels. In particular, transparency coding interacts strongly with luminance and saturation coding and should not be used in conjunction with them at all.

Question 10

Given data that is ordinal and quantative. What kind of colormap would you use to encode this type of data.

Answer 10

A continuous colormap with Luminance encoding the ordinal nature of the data and hue encoding the quantative nature of the data.

Question 11

Given data that is categorical and unordered, what kind of colormap would you use?

Answer 11

Depending on the number of categories, Hue can be used as bins of color representing each category.

Question 12

When using categorical colormaps, what do you have to keep in mind?

Answer 12

A good set of initial choices are fully saturated and easily nameable colors, which are also the opponent color axes: red, blue, green and tellow. Other possibilities when more colors are needed are orange, brown, pink magenta, purple and cyan.

Careful attention must be paid to luminance and saturation.

Question 13

Why are rainbow colormaps not ideal as a type of ordered colormap?

Answer 13

1. Hue is used to indicate order here, despite being an identity channel without an implicit perceptual ordering.
2. The scale is not perceptually linear, steps of the same size at different points in the colormap range are not perceived equally by our eyes.
3. The fine detail cannot be perceived with the hue channel; the luminance channel would be a much better choice, because luminance contrast is required for edge detection the human eye.

Question 14

How would you fix a standard RGB rainbow map?

Answer 14

Design a monotonically increasing luminance colormap instead. Multiple hues are then ordered according to their luminance from lowest to highest. The varying hues then allow easy segmentation into categorical regions.

Question 15

Name two things you can do to make your visualization colorblind friendlier.

Answer 15

1. In addition to hue, also encode saturation alongside it.
2. Be wary of green-red color ramps.
3. Use a color blindness simulator to check.

Question 16

What kind of channel is size and what is it suitable for?

Answer 16

Size is a magnitude channel suitable for ordered data. It interacts with most other channels: when marks are too small, encoding in another channel such as shape or orientation simply cannot be seen. Size interacts particularly strongly with color hue and color saturation. The more dimensions you encode in a size like lines, areas or volumes the lower the accuracy is for humans to read.

Question 17

What kind of channel is angle and what is it suitable for?

Answer 17

The angle channel encodes magnitude information based on the orientation of a mark: the direction that it points. There are wo different wats to consider orientation that are essentially the same channel. With angle, the orientation of one line is judged with respect to another line. With tilt an orientation is judged against the global frame of the display. While this channel is somewhat less accurate than length and position, it is more accurate than area

Question 18

What kind of channel is curvature?

Answer 18

One that you should not use

Question 19

What kind of channel is Shape and what is it suitable for?

Answer 19

Shapes can be applied to point marks or to line marks, but not to area marks. Given that the point mark is large enough you can incorporate many bins in the visualization. However, there is a strong interaction between size and shape. You need to have enough pixels per mark to make this work well. Shape can interfere with other channels in the same way that size coding does, and it can interfere with size coding itself. For example, filled in shapes like disks are a good substrate for also encoding with color hue.

Question 20

What are the upsides and downsides of motion as a visual channel?

Answer 20

Motion is less studied than other visual channels. Motion is extremely salient, and more-over motion is very separable from all other static channels. It is strongly separable from the highly ranked channels typically used for showing groups and categories such as color and spatial positino. The strength and weakness of motion is that is strongly draws attention; it is nearly impossible to ignore. Basically moving or flickering marks are annoying, although very separable. The motion channels are most appropriate for highlighting, where drawing the user’s attention away from the rest of the scene is exactly the goal, particularly when the highlighting is transitory, rather than ongoing.

Question 21

What are univariate and multivariate colormaps?

Answer 21

The safest use of the color channel is to visually encode a single attribute; these colormaps are know as univariate. When encoding two variables with color it is called a bivariate colormap. When one of the two attributes is binary, then it is straightforward to create a comprehensible bivariate colormap with two families of colors by fixing a base set of hues and varying their saturation. When both attributes are categorical with multiple levels, results will be poor, so don’t.

Question 22

Explain sequential colormap.

Answer 22

A sequential colormap ranges from minimum value to a maximum value. If only the luminance channel is used, the result is a grayscale ramp. When incorporating hue, one end of the ramp is a specific hue at full saturation and brightness. If saturation is the variable, the other end is pale or white, when luminance is the varying quantity, the other end is dark or black.

Question 23

Explain a diverging colormap

Answer 23

A diverging colormap has two hues at the endpoints and a neutral color as a midpoint, such as white, gray or black, or a high luminance color such as yellow.