VCU PR Posts Interview With Mapping Color Team

VCU Public Relations has published a series of interviews with the creative team behind the Mapping Color software. You can read the whole article by clicking on the image below.

The color map has both a theoretical cylindrical model of the visible spectrum as well as this unusually shaped model of color in RGB space. Different values of different hues can be represented with greater chromaticity than some others.

The color map has both a theoretical cylindrical model of the visible spectrum as well as this unusually shaped model of color in RGB space. Different values of different hues can be represented with greater chromaticity than some others.


Mapping Color

Although color is a concern of disciplines as diverse as art, science, business and medicine and has experts in industries as diverse as the breakfast cereal industry and the Pentagon, it is arguably one of the most widely misunderstood yet widely used visual principles. This misunderstanding stems from the fact that the vast majority of people believe color is a physical property of an object rather then the perception and interpretation of visible wavelengths that emanate from or are allowed to pass through an object that has been affected by luminance. Although the ability to create and view objects in virtual space (three-dimensions) has been readily available for decades, color systems are still being presented as two-dimensional or as two-dimensional depictions of three-dimensional models.

Even those who use color most -- fundamentally misunderstand many of the basic principles of color theory: where and how color is generated, how it interacts with an object, how a person interprets the incoming light wavelengths to generate a perceived color, and how color impact is produced and manipulated in various media. The absence of any definitive tool that accurately define and articulate the essence of color science and color math may contribute considerably to this problem. 

Figure 1: The Visible Spectrum

Figure 1: The Visible Spectrum

For years students have been taught, that the colors in the visible spectrum are red, orange, yellow, green, blue, indigo, and violet (ROYGBIV) as identified by Sir Isaac Newton in the 17th century, although anyone viewing the spectrum can easily see an infinite array of colors (Figure 1). Further, art students, who are exposed to more sophisticated color theory, are being taught that the primary colors (colors that cannot be mixed from other colors) are red, yellow and blue (subtractive primaries), whereas photography students are taught that the primaries are red, green and blue (additive primaries), as if these two systems are mutually exclusive. A basic problem in teaching such a subtractive system to art students, in particular, is identifying which red, which yellow and which blue are being referenced. Color identification by means of common names (red, blue, etc.) is inadequate for the identification of the spectrum of colors available to us (Figure 2).

 

Figure 2: Where yellow stops and green begins

Figure 2: Where yellow stops and green begins

In professional practice, the tools used to identify, analyze, and manipulate color are more sophisticated but still lack a level of precision and functionality that could benefit a range of fields in which color analysis is important. Currently, individuals working with pigments (graphic, interior and fashion designers, painters, etc.) are using outmoded color tools or are relying on subjective judgments to select colors with little or no understanding of a color’s location in color space[1], and with little or no understanding of how the appearance of a specific color will change as a result of changes in illumination.

 

Color Gamut: An Interactive 3-Dimensional Color Model, has addressed this problem by creating an interactive, web-based software tool that can be used by artists, engineers, scientists, educators and students. The model maps specific colors, color space and all its complements in various formats. The project involves the development of a system to identify colors based on their hue (color), saturation/chroma (purity) and value (lightness/darkness), and to show their location in relation to all other colors using Cartesian coordinates on a 3D interactive and programmable model (Figure 3). To our knowledge, this will be the first interactive and programmable three-dimensional model defining color space available internationally.

 

[1] Color space: The range of light to which the eye is responsive. This allows a variation in value (intensity), in hue (color appearance) and saturation (chroma or how far it is from a grey scale). A color space should allow an observer to consistently evaluate color differences.

 

 

Figure 3: Hue, Chroma and Value

Figure 3: Hue, Chroma and Value

Albert Munsell first proposed the concept of identifying colors by using a three dimensional component model in 1915 (Figure 4). His model was based on a five-color system (yellow, green, blue, purple and red). Since then, a few models have been proposed that are based on the more accurate six color (YRMBCG) system but none are interactive and available for general use. The principles of color theory can be better presented using contemporary technologies. Having the ability to navigate a color model in virtual space is key to understanding color. Currently, color is taught using textbooks written half a century ago. Color mixing is taught using subtractive primaries as a complete system. Many color theory textbooks published in the past ten years do describe a system that combines both the subtractive primaries (CMYK or Blue, Red and Yellow) with the additive primaries (RGB), but remain two dimensional depictions of three dimensional space and cannot be made interactive by virtue of the traditional book format. 

 

Figure 4: The Munsell system

These color picker systems do not take advantage of current technologies, they are primarily two dimensional, and do not take into account Value, Chroma and Hue, VCh, color location. Without a standardized tool developed from available technologies, any industry that requires standardized systems to identify colors functions awkwardly with antiquated tools to produce, select, and identify color. The Color Gamut Project – a collaboration between VCU artists, physicists, computer scientists, and software engineers – will quite literally revolutionize how color can be used in science, information technology, and manufacturing.

 

It may be easiest to describe our model as a “point cloud” or a mass of floating marbles (Figure 5). Each marble identifies the location of a specific color so that the viewers can navigate around and through the model, allowing them to not only see the location of a specific color, but also see its location in relation to all other visible colors. 

Figure 5: Top - Full model          Bottom: RGB gamut space

Figure 5: Top - Full model          Bottom: RGB gamut space

The color model developed for the Color Gamut Project will identify and describe colors based upon human perception, differing substantially from the subtractive and unscientific approaches most commonly employed in education and practice. It utilizes three parameters; the color or hue or wavelength (does it seem to be red, or green or blue or any other general color term); the purity, or saturation or chroma which describes how much of the color appearance can be attributed to a unique hue (or color or wavelength); and the value or lightness, which relates a color to an equivalent achromatic scale. These parameters are the same as those available on instrumentation used to quantify the color of a sample and are commonly used in the pigment and color measuring industries for quality assurance.

In order to provide greater accuracy and specificity, we are proposing a color identification system based on a color’s location on a Cylindrical Cartesian coordinate system, which would use degrees (0 - 360) to identify hue (Figure 6), a scale (0-100+) to identify chorma/saturation (Figure 7), and percent (0%-100%) to identify value/lightness (Figure 8). The model is based upon the CIELab or CIELuv coordinate systems that have been developed by the Commission Internationale de l’Eclairage over the last 80 years to provide a good approximation of the response of the eye to color differences. 

Figure 6:  Using degree to identify hue, as opposed to using common names, allows for unlimited specificity

Figure 6:  Using degree to identify hue, as opposed to using common names, allows for unlimited specificity

Figure 7: Chroma is plotted on the horizontal axis.

Figure 7: Chroma is plotted on the horizontal axis.

Figure 8: Value is plotted on the vertical axis.

Figure 8: Value is plotted on the vertical axis.

If, for example, you were to apply 3D color mapping to Adobe Photoshop’s current color picker, chorma would be aligned with value. It would additionally show the relative distance between full-chorma and the achromatic scale (Figure 9).

Figure 9: top: Adobe Photoshop’s current color picker; middle row: showing chroma relative to value; bottom: granular mode – showing CMYK gamut within RGB gamut.

Figure 9: top: Adobe Photoshop’s current color picker; middle row: showing chroma relative to value; bottom: granular mode – showing CMYK gamut within RGB gamut.

The full range of colors that a person can perceive is known as the human gamut. Any mathematical model used to display color space will most likely either use a screen (each pixel is an additive combination of red, green and blue light sources) or printed material (each point in the image is a combination of cyan, magenta, yellow and black pigments). Both of these display modes have a limited gamut, or range of colors, that can be displayed, and both can only display a fraction of the colors perceivable by humans. It is our intent to use a translucent shell to show the location of colors that are part of the human gamut but which cannot be represented colorimetrically because of the limitations of the display media gamut (Figure 10).

Figure 10: CIE chromaticity diagram showing the RGB and CMYK gamut in relation to full human gamut.

Figure 10: CIE chromaticity diagram showing the RGB and CMYK gamut in relation to full human gamut.

We can now upload an image to our model and the model will display the gamut space (all the colors present) for the image (figure 11).

Figure 11: Color space of the John Henley photo that is pictured on the left.

Figure 11: Color space of the John Henley photo that is pictured on the left.

Additional developments could include a fully interactive three-dimensional model of the visible color spectrum with a host of various controls to manipulate the model. This would allow the user to draw out and highlight various connections and relationships between colors that could generate a host of useful applications yet unimagined. In addition to the visualization of RGB gamut space, we expect the final model to include variation showing CMYK, Pantone, and artist pigment space. An interactive model would allow the user to input any pigment color, or a pallet of pigment colors and not only see individual color location, but also show all the colors that can be mixed with a pallet of colors, for example how color space changes when you substitute Cadmium Red for Alizarin Crimson. The use of commercially available image capturing devices would enable a user to do real time analysis and see how objects were changing color with time or environmental conditions.

Using the image or color description that can now be obtained with color measuring equipment the model could be used to show where sample colors are located with regard to control specifications. Color differences or color sample quality assurance can be done in a 3-dimensional space rather than relying on multiple 2-dimension charts. The model also has potential application in biology where it could detect subtle changes in the color of foliage brought on by environmental conditions, or monitoring the quality of artwork to detect gradual deterioration. It is anticipated that using hand-held image capturing devices, the model could be used to detect real time changes in color. These applications will be evaluated once the model is improved to provide histograms of colors appearing in an image and the computer code is improved to allow more efficient processing of images. Since this model is based on the mathematical identification of color rather then using perceptual differences, color specificity is not limited by the range of common color names. Using mathematical location to identify color will ultimately break down language barriers in specifying colors. (Figures 12a, 12b)

Figure 12a: Each of these circles can be described as “red”, yet each is clearly distinguishable as a different color.

Figure 12a: Each of these circles can be described as “red”, yet each is clearly distinguishable as a different color.

Figure 12b: Using mathematical location instead of the word “red” allows for individual color identification.

Figure 12b: Using mathematical location instead of the word “red” allows for individual color identification.

Key personal

 

Robert Meganck, Chair, Communication Arts, VCU

Additional information can be found at http://meganck.com/

 

Matt Wallin, Associate Professor, Communication Arts, VCU

Additional information can be found at http://mattwallin.com/

 

Dr. Peter Martin, Ph.D, Physics and Mathmatics

David Jackson, Ph.D. candidate in Computer Science at VCU

Brian Glass, Ph. D. candidate in Computer Science at VCU

 

Shaun Graham, undergraduate at VCU in Computer Science

Manyhands Collective, a Richmond VA based multimedia production company with a specialization in web and interactive design.

 

References:

Albers, Josef; Interaction of Color, Yale University Press, 2009

Berns, Roy; Billmeyers and Saltzman’s Principles of Color Technology, John Wiley and Sons, 2000

 Boynton, Robert and Peter Kaiser; Human Color Vision, Optical Society of America, DC, 1996

 Fairchild, Mark; Color Appearance Models, John Wiley and Sons, 2005

Gurney, James; Color and Light: A guide for the Realist Painter, Andrews McMeel Publishing, 2010.

International Commission on Illumination(CIE), Proceedings of the 8th. Session, Paris 1931

Itten, Johannes; Elements of Color, Spon Press, Revised 1990

Itten, Johannes; The Art of Color: The Subjective Experience and Objective Rationale, John Wiley & Sons, Revised 1997

Munsell Color Company; Munsell Book of Color, Baltimore Md, 1929 to date.

Parramon, Jose M.; Color Theory, Watson-Guptill Publications, New York, 1989

Sharma, Abhay; Understanding Color Management, Thompson Delmar Learning, 2004

Turner, Joy; OSA OSA Instrumental in Development of the Uniform Color Scales, Optics and Photonics News, Sept. 1999

 Yot, Richard; Light for Visual Artist: Understanding & Using Light in Art and Design, Laurence King Publishers, 2011

 Zelanski, Paul and Fisher, Mary Pat; Color, sixth edition, Prentice Hall, 2010

In addition the following interactive web sites (physlets) provide insight as to how color is created and the interrelationship of several of the existing models used to describe colors.

http://www.highpoint.edu/~atitus/physlets/blackbody/index.htm show how the electromagnetic radiation from a black body emitter changes with the source temperature and what color it seems to be when interpreted by human vision.

http://www.omsi.info/visit/tech/colormix.cfm  color generation by addition and subtraction, and shows how the final visual appearance depends on the relative intensities of the primary colors (red, blue, green for addition and cyan, magenta and yellow for subtraction).

http://www.cs.rit.edu/!ncs/color/aspectr.html  show how the color appearance to the average human vision is dependent upon the spectral distribution of the incident illumination.

http://www.colorpro.com/info/tools/convert.html#START  shows how a single color is represented in several color appearance models (CIE La*b*, XYZ, xyY, Hunter Lab, and RGB) and displays the color parameters associated with that color in each model.

http://www.cs.rit.edu/~ncs/color/aspaces.html   shows graphic representations of the RGB, HSV, YIQ and CIELab color space models and shows a color swath of any chosen location in any of the color spaces.De Grandis, Luigina; Theory and Use of Color, Prentice-Hall, New Jersey; and Harry N. Abrams, New York, 1984

Scientific 3D Color Analysis of the Two Dresses Meme

Is this dress white and gold or blue and black?

The controversy over the dress color most likely rests in our understanding (or lack there of) of the basics of color perception. Our 3D color mapping software puts an end to the debate. The Mapping Color Alpha is capable of full-color image, video and data mapping and analysis. 

When we use terms like: Blue, Red and Yellow to describe a color we are referring to color hue. Hue is only one of three parameters that describe color. The other two are chroma (purity) and value (lightness/darkness). And when we use a word like “blue”, what does it mean. There are numerous blues: ultramarine, cobalt, cerulean, navy, light blue, dark blue, etc. When we were first taught “our colors”, we were shown a color and told that this is “blue. In all probability we did not all see the same color, but simply identified what we saw with the word “blue”.

Most people believe that color is an inherent property, rather than a property of light. Most of us have read about how Sir Isaac Newton splitting a narrow beam of light into a rainbow of colors know as the visible spectrum by letting it pass through a prism and bending white light into different wavelengths.

The reason we perceive a tomato as red is that when it is lit by white light, the surface of the tomato absorbs the short and medium wavelengths, and reflects the long wavelengths -- those associated with red.

A tomato under warm light looks red, as does a tomato under cool light. But when displayed side by side we can see that they are very different reds.  

So the answer to the controversial dress will be found in the light source not in the dress itself.

If we use our model to analyze a red dress and a green dress we show that they occupy two very different color spaces.

Color space of a red dress in 3D space.

Color space of a red dress in 3D space.

Color space of a green dress in 3D space

Color space of a green dress in 3D space

If we take the two photos of the controversial dress and upload them into our color model we see how the color space of each of the images is different.

When viewing our model from the top -- showing hue and chroma we notice that both dress favor a hue axis that runs from approximately 105° and 185°. However the blue/black dress occupies more blue space.

 

Color space of white/gold dress view from top in 3D.

Color space of white/gold dress view from top in 3D.

Color space of blue/black dress view from top in 3D.

Color space of blue/black dress view from top in 3D.

A greater dissimilarity is visible if we see the model for the side views (showing chroma and value).

 

Color space of white/gold dress view from the side in 3D.

Color space of white/gold dress view from the side in 3D.

Color space of blue/black dress view from the side in 3D.

Color space of blue/black dress view from the side in 3D.

The shifts in value, chroma and hue (VCh) is very noticeable when viewing the model in 3D.

Color space of white/gold dress viewed in 3D.

Color space of white/gold dress viewed in 3D.

Color space of blue/black dress viewed in 3D.

Color space of blue/black dress viewed in 3D.

Its worth noting that this analysis was done using photos of the dress and not from viewing the actual physical dress. However, the answer to this dilemma will be found more in the color of light than in the color of the dress. 

Mapping Color Alpha 1.0

The Mapping Color team has been hard at work for the last several months and are nearing completion of the first Alpha version of the application. Currently the system can analyze images for approximate or precise color location data, input data for mapping, output data results for use in user pipelines, display McAdam's limits, locate pigments, display theoretical color spaces and work within set tolerances for quality control applications. 

We're hoping to have real-time video analysis working within the next few months. There may be a limited opportunity for beta testing in the Spring of 2015. If interested please contact the team via email. Please include a detailed description of your current research/commercial work and how you'd like to use our app. 


Mapping Color presenting at US Patent Office

Robert Meganck, Matt Wallin & Peter Martin presented their 3D Color software prototype today at the US Patent Office as part of the US Department of Commerce i6 Virginia Ventures Challenge. The software is in beta and moving towards a release candidate. The current version of the tool allows for multiple color modeling, analysis and transformation as well as image color analysis and manipulation. 


Mapping Color Project Receives 40K Grant

Chair of VCU Communication Arts Robert Meganck and Communication Arts Associate Professor Matt Wallin have received a $40,000 grant from the Virginia Innovation Partnership i6 Challenge competition for the finalization of their 3D VCH Color Model analysis software project.

In the coming months we hope to have a full working beta of the Mapping Color software tool for desktop and laptop computers. Check back here from time to time for updates!

TEDx RVA

Professor Robert Meganck, prolific illustrator, author, consultant and digital pioneer, believes creative problem-solving thrives when we're willing to make mistakes, re-evaluate possibilities, and most of all, have fun. Associate Professor Matt Wallin has worked around the world as a visual effects artist for more than 15 years. His portfolio of work spans numerous feature films and fine art projects, and he has worked along such notable figures in film as George Lucas, Matthew Barney, Steven Spielberg and Woody Allen. 

Together with their peer Dr. Peter Martin, a scientific consultant and adjunct faculty of the VCU Physics department, Robert and Matt are developing a revolutionary digital platform for re-imaging and understanding color — their project will present color in 3D. The design will reflect our dynamic natural experience of color, with variations that reflect movement, light and density.

In the spirit of ideas worth spreading, TEDx is a program of local, self-organized events that bring people together to share a TED-like experience. At a TEDx event, TEDTalks video and live speakers combine to spark deep discussion and connection in a small group. These local, self-organized events are branded TEDx, where x = independently organized TED event. The TED Conference provides general guidance for the TEDx program, but individual TEDx events are self-organized.* (*Subject to certain rules and regulations)

 

Mapping Color at TEDx - Richmond, VA

This coming Friday, March 22nd, Robert Meganck, Matt Wallin and Peter Martin will be giving a talk at TEDx - Richmond. The talk centers on our research into color science and the development of our real-time three-dimensional color analysis software toolset.

Tickets are still available from the TEDx RVA site which you can visit by clicking the image below. A live stream of the event and the talk will appear on the TED website and mobile app shortly thereafter.