Introduction
Simultaneous color contrast is a known phenomenon; humans perceive the same
physical color differently depending on its surrounding color, or
background. This color difference caused by simultaneous contrast is
particularly problematic in information graphics where colors convey
meanings.
Figure 1. Simple example of a simultaneous contrast effect. The two central
circles share the same physical color.
In traditional graphic design, the simultaneous color contrast effect has
been known by designers for a long time [1]. In general, it is recommended
to select colors which minimize the simultaneous contrast [7][8]. However,
visual designers sometimes utilize the effect to achieve more complex color
harmonies, or to increase the number of colors without adding more inks.
Map design is an example which often takes advantage of the effect to give
the illusion of having more colors than it is actually using [8]. The
effect can also be manually adjusted. For example, when a corporate logo is
used on various colorful photographs, the corporate color may be perceived
differently because of the effect. In such a case, colors can be manually
adjusted by a designer.
The use of computer-based information graphics has raised new problems
caused by the simultaneous contrast effect. First, as more non-professional
people start to use color graphics in a wide variety of software, such as
drawing and chart-making, the simultaneous contrast effect becomes a
problem for the design of effective communication. Second, on a
computer-based dynamic display, such as weather and air traffic, since
background color or position of a graphical element is difficult to predict
at run-time, the simultaneous contrast effect becomes a problem for
reliable communication, or a limitation for color selection.
This paper presents a prototype of a graphics system which allows users to
select colors based on how they are perceived, instead of their physical
specification (i.e., RGB). The system automatically adjusts the physical
color of each graphical element so that all elements that are intended to
appear the same color are perceived that way.
INTERACTION SCHEME
The proposed graphic system considers the color adjustment as an extension
of a color palette. Figure 2 shows a simplified interaction diagram which
compares the proposed adaptive color graphic system with a typical system,
such as drawing and painting. A user selects a reference color with a
background color. The background color can be either chosen from a palette
or from a document itself. While typical graphic systems simply use the
physical component (i.e., RGB) in a document, the proposed approach uses
the automatic adjustment module to maintain the color appearance of the
display target so that it is perceived the same way as the reference target.
Figure 2. Interaction model with and without an automatic adjustment
module.
ADJUSTMENT METHOD
The simultaneous contrast effect has been known for a long time and there
have been a considerable numbers of psychological studies made since the
nineteenth century. After a series of seminal experiments by Jameson and
Hurvich in the 1960's, there has also been research based on the
neurophysiological findings; and a number of models have been proposed and
are being examined.
In this project, I have implemented software which automatically adjusts
color differences caused by simultaneous color contrast based on Jameson
and Hurvich's research [5][6]. First, the system converts three primary
colors (i.e., red, green, blue) on a computer display into three opponent
primary colors (red-greenness, yellow-blueness, and brightness) [3][4].
Then, the appearance color of the reference target is computed by taking
the reference background in to consideration. Finally, the physical color
of each display target is computed as a function of its background and the
appearance of the reference target. In this method, achromatic and
chromatic effects are computed separately, but not independently. The
detail of the adjustment method is described elsewhere [2].
EXAMPLES
Three application examples are selected to illustrate the use of the
automatic adjustment technique. The first is a simulated dynamic display of
butterfly migration, in which locations where butterfly are discovered are
indicated by squares (Figure 3). Geography is shown by using a satellite
map in which the color varies from desert (pale yellow) to ocean (deep
blue) to forest (green). In this display, since the background colors of
design elements may arbitrarily vary, the simultaneous contrast effect
becomes unpredictable. The automatic adjustment improves the display in
three ways. It corrects the perceived size differences of individual
squares caused by achromatic contrast. The perception of depth caused by
this size difference is also adjusted and the squares are perceived as
displayed in a flat plane. The adjustment also emphasizes the perception of
the square as a group.
Figure 3. A display of butterfly migration.
(Note: The result of the automatic adjustment in Figure 3 ~5 are not well reproduced
because of the image scaling.)
The second example is a dynamic display of on-line news which presents
headlines according to location and the news stories are displayed upon
request on the same map (Figure 4). When a story is displayed over a
varying background color, it is not easy to find a right color which does
not have the simultaneous contrast problem. No matter what color is
selected, some part of the text is effected. In this example, color of the
text is adjusted character by character, and the adjustment significantly
improves the readability of the text.
Figure 4. An on-line news display presenting news article based on
their location.
The third example is a typical business graphic. Figure 5 shows a graph
of pedestrian traffic in a train station. The graph is intended for the
comparison of people entering and exiting the station according to the time
of day. Blue and pale yellow are used for the background to differentiate
between morning/evening and during the day. Without the adjustment, the
colors of the bars displayed on different backgrounds are perceived
differently. The adjustment corrected the problem of perceived thickness
differences among bars, similar to the size differences described above. It
also improved the sense of transition from one bar to another; in other
words, the visual flow of each data set.
Figure 5. A bar graph showing pedestrian traffic in a traion station.
CONCLUSION
This paper presents a prototype of a color adaptive graphic system, along
with an interaction model for selecting color based on its appearance. A
computational technique for adjusting simultaneous contrast effect is
briefly introduced. The purpose of this research is to enhance the quality
of visual design by adjusting the simultaneous color contrast effect. The
results of information graphics experiments shows that the automatic
adjustment improves the visual design of information display, as well as
the flexibility of the color choice.
Future research will include further refinement of the automatic adjustment
module, adjustment of arbitrary forms (current system only support text and
rectangles), and study concerning the relationship between simultaneous
contrast and rapid motion.
ACKNOWLEDGEMENT
This research has been done at the Visible Language Workshop, Media Laboratory, Massachusetts Institute
of Technology. I thank members of the VLW for their help. The world map data was provided by Tom Van
Sant of Geosphere, Inc. This work was sponsored in part by ARPA, NYNEX, Alenia, and JNIDS.
References
-
Albers J. nteraction of color. Yale University Press, 1971.
-
Ishizaki, S. Adjusting simultaneous contrast for dynamic information
display. Proceedings of IS&T and SID's Color Imaging Conference,
Scottsdale, 1994: pp137-140.
-
Jameson D. and Hurvich L. M., Some quantitative aspects of an
Opponent-Colors Theory.II. Brightness, saturation, and hue in normal and
dichromatic vision, J. Opt. Soc. Am. Vol.45 No.8, 1955: pp602-616.
-
Jameson D. and Hurvich L. M., Some quantitative aspects of an
Opponent-Colors Theory.IV. Psychological Color Specification System, J.
Opt. Soc. Am. Vol.46 No.6, 1956: pp416-421.
-
Jameson D. and Hurvich L. M., Opponent chromatic induction: Experimental
evaluation and theoretical account, J. Opt. Soc. Am. Vol.51 No.4, 1961:
pp46-53.
-
Jameson D. and Hurvich L. M., Theory of brightness and color contrast in
human vision, Vision Res. 4, 1964: pp135-154.
-
Marcus A., Graphic Design for Electronic Documents and User Interface, ACM
Press, 1991: pp77-96.
-
Tufte E., Envisioning Information, Graphics Press, 1989: pp81-95.
