Question: True or false: When we see the gorgeous, iconic images from the Hubble Space Telescope or the stunning panoramas from the Mars Exploration Rovers, those pictures represent what human eyes would see if they observed those vistas first hand.
Answer: For the Hubble, mostly false. For the rovers, mostly true, as the rovers provide a combination of so-called “true” and “false” color images. But, it turns out, the term “true color” is a bit controversial, and many involved the field of extraterrestrial imaging are not very fond of it.
“We actually try to avoid the term true color’ because nobody really knows precisely what the truth’ is on Mars,” said Jim Bell, the lead scientist for the Pancam color imaging system on the Mars Exploration Rovers (MER). In fact, Bell pointed out, on Mars, as well as Earth, color changes all the time: whether it’s cloudy or clear, the sun is high or low, or if there are variations in how much dust is in the atmosphere. “Colors change from moment to moment. It’s a dynamic thing. We try not to draw the line that hard by saying this is the truth!'”
Bell likes to use the term “approximate true color” because the MER panoramic camera images are estimates of what humans would see if they were on Mars. Other colleagues, Bell said, use “natural color.”
Zolt Levay of the Space Telescope Science Institute produces images from the Hubble Space Telescope. For the prepared Hubble images, Levay prefers the term “representative color.”
“The colors in Hubble images are neither true’ colors nor false’ colors, but usually are representative of the physical processes underlying the subjects of the images,” he said. “They are a way to represent in a single image as much information as possible that’s available in the data.”
True color would be an attempt to reproduce visually accurate color. False color, on the other hand, is an arbitrary selection of colors to represent some characteristic in the image, such as chemical composition, velocity, or distance. Additionally, by definition, any infrared or ultraviolet image would need to be represented with “false color” since those wavelengths are invisible to humans.
The cameras on Hubble and MER do not take color pictures, however. Color images from both spacecraft are assembled from separate black & white images taken through color filters. For one image, the spacecrafts have to take three pictures, usually through a red, a green, and a blue filter and then each of those photos gets downlinked to Earth. They are then combined with software into a color image. This happens automatically inside off-the-shelf color cameras that we use here on Earth. But the MER Pancams have 8 different color filters while Hubble has almost 40, ranging from ultraviolet (“bluer” than our eyes can see,) through the visible spectrum, to infrared (“redder” than what is visible to humans.) This gives the imaging teams infinitely more flexibility and sometimes, artistic license. Depending on which filters are used, the color can be closer or farther from “reality.”
In the case of the Hubble, Levay explained, the images are further adjusted to boost contrast and tweak colors and brightness to emphasize certain features of the image or to make a more pleasing picture.
But when the MER Pancam team wants to produce an image that shows what a human standing on Mars would see, how do they get the right colors? The rovers both have a tool on board know as the MarsDial which has been used as an educational project about sundials. “But its real job is a calibration target,” said Bell. “It has grayscale rings on it with color chips in the corners. We measured them very accurately and took pictures of them before launch and so we know what the colors and different shades of grey are.”
One of the first pictures taken by the rovers was of the MarsDial. “We take a picture of the MarsDial and calibrate it and process it through our software,” said Bell. “If it comes out looking like we know it should, then we have great confidence in our ability to point the camera somewhere else, take a picture, do the same process and that those colors will be right, too.”
Hubble can also produce color-calibrated images. It’s “UniverseDial” would be standard stars and lamps within the cameras whose brightness and color are known very accurately. However, Hubble’s mission is not to produce images that faithfully reproduce colors. “For one thing that is somewhat meaningless in the case of most of the images,” said Levay, “since we generally couldn’t see these objects anyway because they are so faint, and our eyes react differently to colors of very faint light.” But the most important goal of Hubble is produce images that convey as much scientific information as possible.
The rover Pancams do this as well. “It turns out there is a whole variety of iron-bearing minerals that have different color response at infrared wavelengths that the camera is sensitive to,” said Bell, “so we can make very garish, kind of Andy Warhol-like false color pictures.” Bell added that these images serve double duty in that they provide scientific information, plus the public really enjoys the images.
And so, in both Hubble and MER, color is used as a tool, to either enhance an object’s detail or to visualize what otherwise could not be seen by the human eye. Without false color, our eyes would never see (and we would never know) what ionized gases make up a nebula, for example, or what iron-bearing minerals lie on the surface of Mars.
As for “true color,” there’s a large academic and scholarly community that studies color in areas such as the paint industry that sometimes gets upset when the term “true color” is used by the astronomical imaging group. “They have a well-established framework for what is true color, and how they quantify color. But we’re not really working within that framework at that level. So we try to steer away from using the term true color’.”
Levay noted that no color reproduction can be 100% accurate because of differences in technology between film and digital photography, printing techniques, or even different settings on a computer screen. Additionally, there are variations in how different people perceive color.
Bell concluded, “What we’re doing on Mars is really just an estimate, it’s our best guess using our knowledge of the cameras with the calibration target. But whether it is absolutely 100% true, I think its going to take people going there to find that out.”
For more information see http://hubblesite.org/ or check out Jim Bell’s 2006 book “Postcards From Mars.”