The art of seeing...colors in the universe are "private"

The intention of this overview is to simply help understand how we humans perceive and finally understand the colors of the world we are born into.

I have witnessed many discussions about the "true colors in LRGB- and DSLR-astrophotography" that repeatedly have taken place all over the globe in different forums at several occasions and frequently yield a result which clearly misses the punch-line of human color-sensation. The audience of such discussion sincerely longs for "objective colors" at all costs, negleciting some very important facts I'd like to bring to your consciousness. Certainly we have to try our very best to avoid picking up color artefacts by all sorts of gradients, filter-diversities, atmospheric aberration, optical indices of the image trainwe use and many more. Nevertheless each image train is totally different, each location we acquire the image-data is different. You might have attended some astro-imaging conference where a couple of well renowned imagers have presented their photos of maybe the same object and claimed to have gotten the "right" colors. (Google for some images of one and the same object: but be prepared for the high jump) And has there not been a huge difference in the appearance of such photos? Naturally! Because the the most important factor in color-composition is us. I hope I can make my point clear and bring some light (hopefully colorful) into this fascinating topic. (False color imaging is not affected by these thoughts, as in false-colos are deliberately used to enhance and emphasize contrast, and do not claim to show the object in "true-colors,...hence the term "false-colors"

Let us first start by looking at some essential features in the human "optical train", which we have to be aware of before we discuss what happens from the entrance of a photon until the individual interprets it successfully as an image of the world he lives in.

The features that represent the classical optical elements (can be looked up in books and are not of interest here) cast the image onto some specific retinal cells which are widely and dramatically underestimated for their competence in the "act of seeing".

We will not talk about cones and rods very much here, as general knowledge educates about them, but we will pay more attention to a specific brand of retinal cells, which are not so commonly known, but deserve more respect and attention. Cones would act as the color-sensors (Red, Green and Blue), while rods would make the "luminance sensors". Let us take a closer look onto the retina and learn about its architectural structure next.

This would be an image,an optician gets to see when investigating the "background" of the eye-ball. From the right comes the blood-supply, which runs in the center of the optical nerve; this spot is "blind", as there are basically no light-dependant sensors. By the pattern of distribution of the vessels you can tell that there must be a spot of certain importance (red) - the fovea centralis (f.c.) , which indicates the location of the highest density of cones.This spot aka "yellow spot" holds a couple millions of rods, but almost no cones. However densitiy of cones is much higher in a circular ring-shape field around the f.c. See images right below...

Illustration of densitiy of the photo-receptors in a human retina. Interestingly enough, the blue receptors are the most recent evolution in color-sensation. They are thought to have come into existence only some hundreds of thousand years ago. Typically the average retina holds about 12% of such blue-receptors (of all RGB), while red and green sensors show great variation in dispersion.

An average human eye hosts about 6 Mio cones and 120 Mio rods. The appearance of cones is dramatically high on at the f.c. and extremly lowat the periphery of this particular spot. The optical focus is found on the f.c. So whenever you concentrate on a particular object of interest, millions of cones are activated and will provide a sharp image, as long as the light-densitiy is high enough. When it becomes dark, cones will cease from being active and rods will take over the task of seeing, as rods are 100 times more sensitive to light than cones are. However, the density of rods is extremly low in the f.c. (as cones occupy all the place of that retinal area - this shows that we are designed to be "day-active" creatures, as the very best optical image is produced at bright day-light).

Cones and Rods

When we bring the typical pattern of distribution of cones and rods as well as the density of these cells to our mind, it should become evident, why we cannot see sharp images at night and why the image is in grey-scale. (at night all cats are grey...) In regards to astronomy we now understand, why we have to observe pale astronomical objects "indirectly" in an eyepiece: because usually the light-densitiy is insufficient to activate the cones. Rods density is much higher around the f.v. (where there is literally no space for rods in favour for the RGB image-sensor-facility). So much for cones and rods. Now let us draw our attention to the real venerated heroes in the retina - the socalled "interlaced cells".

There is a considerable large variety of these interlaced cells which are put in between the cones, rods and the optical nerve-fibre which comprises the optical nerve.

What is the task of these little guys? The short answer is: without their function, cones and rods would not yield an image at all.

From right to left the order of retinal structures would be: optical nerve-fibre interlaced cells cones and rods base-membrane.

In this graph an assumed light ray would penetrate the retina from right to left. "Behind" the photoreceptors a pigment cell layer operates as a kind of reflector. In some mammals this very cell-layer is enormously effective and could be regarded as "rest-light amplifier", which for example allow cats to see pretty well in the dark, whereas humans would get lost easily. What is the job of the interlaced cells?

The part of the human brain which is occupied with intellectual operations definitely does not have the time to investigate the incoming information of the eyes. Computing time is "expensive" and so the brain has shifted authority to the human eye, respectively the retina. Seen from an ontogenetic view the retina actually is a part of the human brain. During the very first weeks in an embryological development the retina literally gets pinched off from the brain and migrates to the front part of the forming head. An "umbilical cord" provides aconnection to the brain, which we call the optical nerve. Any information that runs from the eye passes through the optical nerve when "radiating" into the brain. The retina could therefore be described as sort of "secretary of state", as it has been equipped with competence and authority in image-processing and forming a pre-processed image.

The mission of the interlaced cells could be described as "pre-processing": While photoreceptors can no more than be "active or inactive" in a "binary" way (on - off) these cells analyse the incoming information from the photoreceptors for: saturation of RGB, hue and brightness. The optical image that is cast onto the retina by the optical features of the human eye is replete with chromatic aberrations, as we only have one single lens in the optical pathway. Certainly cornea, both liquid eye-chambers and the glass-body have different optical indices but still, the image is not perfectly sharp and it is not "apochromatic". Also of course, the comatic aberration of a single-lens-system has to be compensated for - this is performed by the fact that the retina has a spherical shape. Nevertheless, the "raw-image" needs to be preprocessed. And this pre-processing is done by the interlaced cells: What we astro-photographers usually do with our stacked images is a deconvolution to enhance contrast and bring out more subtle features of the objects we have imaged. Frankly, a sort of deconvolution is performed by these interlaced cells! This is awe-inspiring!

As one eye holds about 125 million retinal sensors, and every single photoreceptor sends out one nerve-fibre, an optical nerve would be as thick as an arm if the interlaced cells would not pre-processed in terms of data-rejection and therefore data- reduction. By means of this, all 125 million fibres can be reduced to 1 million nerve.fibres (in average) channelling the final product of pre-processing to the brain. Condensing this information about the interlaced cells, we should humbly stand before these guys and be grateful for what they do for understanding the world we live in: Contrast enhancement, deconvolution, data-rejection and data-reduction, RGB-preprocessing in terms of analysis for hue and saturation, just to name the very essential parts.

Optical information passes through many parts of the brain on its way to the occipital region of the brain, where a significant part of image-assessment is performed. Coming right ouf of the orbital zone, the nerve is associated to the olfactory tract...
...it also passes through areas of the brain that are competent for linguistic understanding, creation of individual sounds and language, in order to be passed on to regions of the brain that deal with the creation of emotions and feelings...
...finally the information from the "secretary of state" touches down in the occipital area of the brain, which truly is a work-horse in image-processing. The image is analysed for vertical, horizontal and diagonal features, motion is being tracked here, as well as typical frequently appearing shapes are identified. But the occipital area aka "optical brain" is connected to many more associated brain-stations...
...as multiple so called "radiations" of the optical fibres are then projected to almost all over the hemispheres. An image must not only be identified for spacial implications, color and shape - but the information must be understood in order to make sense. What is important or unimportant to the individual,´? A huge data-base, the memory of a person is crucial to "re-evaluate" and simply remember persons, things, etc...
...we also know about a significant discrepancy of color-awareness in male and female. In addition to that complexity recent studies reveal that the right and left hemisphere of the brain of an individual "sees" the world through - literally - different eyes, as the information gets rated very diversely. The so called smallest range of alertness seen over time is about 1/18 of a second. Isn't it amazing that 18 times a second an image is cast on the retina, preprocessed, and postprocessed in the brain, coordinated amd understood? And isn't it even more amazing how far outinto space we can reach with these facilities? Given a dark sky (without any light-pollution) you can easily see stars which are more than 1000 light-years away, and you can even spot galaxies that are millions of light-years away in outer space.

A huge, fast and ministrant memory- bank is an inevitable factor in understanding the world, that is detected by the optical and nerve structures associated with the act of seeing. This memory bank starts to be filled up with data even before birth is given to the individual. From then on everyone of us had to learn to see. A ball, a cat, a voice, a table had to become what it is commonly held for. Particularly sensation and understanding colors is a very complex and sophisticated process. Colors may have a "gate-keeper" function to the individual as they serve as warning-signals for example, and this is very variable from one culture to the other. Also discrimination of colors is very personal, as we get taught (by persons and environment) very differently in our lives. What I sense as a specific "red-tone" for example most decidedly is not the same "red-tone" another individual would experience, especially if the other individual is female. In every-day-life, the distinction of subtle color-tones is not noticeable. But when it comes to astronomical image processing, subtle distinctions can not be identified in the same way, as we simply have to accept, that to see is a skillful and very personal act of art. Knowing that the density and distribution pattern underlies a great variation and therefore color-composition (specifically R, G) is also personal, should let the idea of "individual color-experience"and as a consequence the impossibility of seeing "true" colors or "objective" colors stand for what it is: A fact. Any discussion of such virtual surrealities does not make much sense as seen from the physiological and socio-cultural standpoint of status.

What consequence does this have? That baldly depends on you! Hopefully no more than to undauntedly realize the necessity to accept that all we can achieve with our humble astronomical image-processing is an interpretation of the cosmic universe, to which we are all an individual colorful blossom.

Dietmar Hager M.D., F.R.A.S. (IV-2009)