Continuing on my photographing color theme from the last color post, we have green. I don’t quite know where these posts are going, other than photographic exercises in seeing color in the world around us, but they may turn into a series of riffs on color as its beginning to tickle some thoughts on camera sensor design and biologically informed computing. Perhaps thinking about color is the right place to start all that complexity?
Given that the last post was about yellow and blue, its appropriate to note that green pigments were originally created by combining the yellow and blue pigments. Green pigments were originally created by grinding up minerals like malachite until approximately 1500 followed by newer copper based pigments like Paris Green that were, like many early synthetic pigments, quite toxic. Fortunately, many of the more modern green pigments are far less toxic.
These days, so much of imaging is digital in nature. I’ve been thinking quite a bit about camera sensors lately and how they were originally designed to mimic how engineers thought the human eye “saw” things. It turns out that like most things, biology is a bit more complicated than we originally thought and the one color that I most strongly associate with biology is green. Green is the color of life. The transitory color of leaves in-between their birth and eventual death and the color of paradise. Its a color that is starting to fade from Northern hemispheres this time of year and while I am glad to have the yellows, oranges, reds and browns in the change of seasons, I’ll miss the green. Green really starts to be missed around January, after the holidays when the world around me has become white, brown and black for a few months. I suppose that it is no wonder that we miss green when evolution has engineered two of our four pigments in the photoreceptors of our eyes (rod and cone) to see middle wavelength hues, centered at 530nm for “green” opsin and 496nm for our rod opsins. Even the long wavelength “red” cone absorption spectra extends well into the green wavelengths. This says something about the importance of being able to recognize green to our survival.
Color is an illusion… Its a fabrication of our brain to interpret energy and wavelength that may or may not be perceived the same way by the person next to you and certainly not in the same way as another species. Because birds see four spectral colors, they likely see a different green than we do particularly as they have oil droplets that act as additional spectral filters on their photoreceptors. Turtles see more colors than do birds, so their perception of green is likely even more different. In fact, even between humans who are not color blind, there are different ratios and distributions of cone photoreceptors for red, green and blue, yet psychophysics experiments show that we all agree on what “green” is and approximately where in color space it resides. Because green is the pigment found in chlorophyll, most species would likely come to an agreement on what “green” is as well.
From a camera design perspective though… green is an interesting proposition. Camera sensors are typically designed with equal ratios of red, green and blue color imaging sites ála the Bayer filter. Other camera manufacturers like Fuji have taken a slightly different approach with their EXR design and the X-Trans sensor design. Notably, the X-Trans design increases randomness of the pixels, but appears to favor statistically more green pixels so that red, green and blue pixels can be physically approximated to reduce “false colors”. I’d be curious to find out what the actual ratios are. Fundamentally though, thinking about “green”, I am wondering if there are more biological lessons that could be introduced into camera sensors by paying attention to how millions of years of evolution have crafted biological imaging sensors… That however, is another post.
These are the fundamental colors seen in the above image reduced to 8-bits.