Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Future Human: Making human's invisible without a suit

  1. Oct 17, 2006 #1
    Greetings, in the physics forum we've had a discussion on how to make a person invisible. Many people suggested complicated invisibility suits which I won't get into. I had an idea of how we could possibly biologically engineer a human body to be invisible without the need for a suit. I'm hoping you guys can give some input into its plausibility. Its a pretty fun read, enjoy!


    Most color as we know it arises from the properties of molecules when their Carbon backbone structure has 'conjugated dienes'. This means in lay terms that there are Carbon atoms strung together with alternating single and double bonds (ie. -C=C-C=C-C=C- etc.). The molecule in Rods and Cones that allows us to 'see' light is also a conjugated diene, called 'cis-11-Retinal'. When a photon strikes 'cis-11-Retinal' it makes a conformational change (changes shape) into 'trans-11-Retinal' and essentially opens up a gate through which a tiny electrical current can now pass through and eventually reach the brain, where you perceive this signal as light.

    However, 'cis-11-Retinal' is always imbedded in an even larger molecule which is called 'Rhodopsin'. Rhodopsin is a large globular protein (globular basically means spherical) which always has the same basic structure, but there are a variety of Rhodopsin proteins which all vary slightly with the addition of a few atoms here and a few atoms there. Its these small variations that make all the difference to you and I when we perceive the wide variety of colors that we can perceive. For only a specific type of Rhodopsin will allow incident (incoming) light to pass through it and strike the all-important 'cis-11-Retinal' whereupon whenceforth the electrical signal is passed onwards up to the brain.

    That is why our 3 color sensing cones, namely the L-cone (Red), M-cone (Green) and S-cone (Blue) are only good at sensing the color of light that their specific 'Rhodopsin' protein lets through.

    With all of this background in mind, consider this possible solution for biological invisibility without the need for a suit:

    It is possible in theory that you could line your body with Rhodopsin proteins all along the epidermis of your skin and you could then use these proteins to detect all the incoming light around you from all directions (the INPUT part of the problem). Now we come to the more difficult problem of how to get the body to OUTPUT the colors of light we need to perfectly mimic the INPUT signals we have just received.

    Generating the proper OUTPUT of colors all over your body surface would require each skin 'pixel' to be a maximum size of say 0.1 mm to ensure that no human eye could be visual acute enough (have good enough vision) to see that you are not invisible, even when their proximity to you is within a foot. Recalling that molecules emit color based on their 'conjugated diene' structure, it is forseeable that we could genetically engineer proteins that produce molecules of all the required colors for our OUTPUT. These proteins could be place locally in the epidermis where they would be interspersed in a mosaic beside our INPUT 'Rhodopsins' and then an INPUT 'Rhodopsin' situated 180 degrees (opposite side of the body) could send an electrical signal through a nervous system (a special nervous system suited to connect INPUTS and OUTPUTS) to tell the OUTPUT protein to generate a molecule that's color mimics the 180 degree INPUT.

    This is my idea of how to create biologically induced invisibility. There are several little quarks that I can think of that would need to be smoothed out, for example since the body is not perfectly spherical, the INPUTS and OUTPUTS would not really be perfectly coupled by 180 degrees. Just think of a woman's breast and her back, clearly the INPUT 180 degrees from her breast would not correspond to the correct mimicry of her OUTPUT on her breast, since her breast is curved and her back flat. But I think that you can appreciate that this problem could be worked through by finding the proper INPUT/ OUTPUT mimicry angles needed for all of your body surfaces and its skin pixels.

    Other than that, there are a handful of other hurdles to overcome, however I think the guts of my solution is pretty good.

    What do you think?
  2. jcsd
  3. Oct 29, 2006 #2
    56 views and not one response? Surely we can do better than that people. Criticism, suggestions and comments WELCOME!
  4. Oct 29, 2006 #3


    User Avatar
    Gold Member

    How can one pixel output the correct colour as seen from more than one angle?

    If I am standing in front of the target, I might see green forest behind their nose. If I move one foot to the left, I might see a red car behind their nose. How can a single pixel "know" where I am looking from? In fact, since I have stereo vision, I'm looking from two places simultaneously.
  5. Oct 30, 2006 #4
    You make a good point. The system would have to know what angle the viewer is looking from. If there were two viewers looking at the suit from angles seperated by 90 degrees, the suite would only be able to fool one of them. BUT, if the pixels were of an odd 3D shape like, say a dodecahedron, each planar surface of the 3D shape could have its own color. This could perhaps fool more than one viewer, but not many more, I admit.

    As for stereo vision, the angle of seperation between your eyes is so minimal that I don't see this as a legitimate concern unless of course, you were standing just inches away from the suit, where such an angle of seperation translates into a fairly large distance relative to the pixel sizes.

    FInally, determining what angular direction the suit should play 'chameleon' to would require detecting a person's line of sight, which isn't too difficult given current retinal detection technologies. IF the suit were to fool digital eyes, like video cameras, it would need to detect the special glasses used in their lenses to determine their respective lines of sight. Both are very doable given even current technologies.

    Still the scenario you implied about having multiple viewers from multiple angles is a very valid criticism and my response, equally valid, falls very short of being able to fool more than perhaps 3 or 4 viewers. I'm open to your ideas Dave. And thanks for you insights.
  6. Oct 31, 2006 #5


    User Avatar
    Gold Member

    But each viewer would be able to see more than one plane of the dodecahedron. In fact, the one plane of the ddh that is perpendicular to their line-of-sight would be only a fraction of what they would see.

    Try stacking a few together and then colour only one face of each. You'll see a patch of colour amongst mostly uncoloured tiles.

    It would first have to recognize that there is a person or persons looking on. It would have to know that "there's someone standing over there in the doorway".

    Hm. Well, it wouldn't really have to know where the camera is looking, as long as it knew there is a camera, it can play "better safe than sorry". So that's a point in your favour.

    What level of chameleon are you looking for? Are you looking for invisibility? Or are you looking for camouflage? I could see camouflage, but invisibility is orders of magnitude harder.
  7. Oct 31, 2006 #6
    Ok, I disagree with all of your last criticisms. I'll just speak to one that I think needs answering. The Dodecahedron scenario you imply refers to pigment colors like that of a rubick's cube. Obviously my solution would use an optical output, not pigments and the optical output could be thought of as fibre optics or lasers if you like. These are highly directional and you cannot see them unless your line of sight is directly (or very nearly directly) in line with them. If we were talking about pigment colors you would be correct.
    Last edited: Oct 31, 2006
  8. Nov 1, 2006 #7


    User Avatar
    Gold Member

    Oh I see. You're not talking about selectively outputting colour based on who's looking, you're talking about continually outputting colour in the right directions at all times. So, the system wouldn't have to know who's looking.

    Doesn't that sort of mean you're got an artifical (even if biologically-based) layer of input and output (lasers)? And doesn't that go back to the "suit" idea?

    I guess I have to read your initial idea more carefully; it does seeem that you were suggesting that skin "pixels" produce pigments.
  9. Nov 1, 2006 #8


    User Avatar
    Gold Member

    great theory. i encourage looking into it further.
    but i have to comment. its very complicated
    always renember KISS
    keep it simple stupid.
    no accusations on the stupid part.

    this will be an extreemly large system
  10. Nov 1, 2006 #9


    User Avatar
    Gold Member

    That's what they said about cellbricks (I mean cellphones) two decades ago.
  11. Nov 2, 2006 #10
    i was thinking about this in a mechanical way to make it and it would be a suit but i just wanted to share, that you could have micro pixel cameras attached to one side of a suit and a reciving tiny screen on the other side of yoru body then the image infront of the person would be projected to their back pixel screens making them blend in, would it work? and do we have the technology to do it is my question?
  12. Nov 2, 2006 #11


    User Avatar
    Gold Member

    It has in fact been done.
    http://projects.star.t.u-tokyo.ac.jp/projects/MEDIA/xv/oc.html [Broken]
    Last edited by a moderator: May 2, 2017
  13. Nov 2, 2006 #12
    haha cool thanks for the site

    EDIT: But wait this is with a whole projection screen on the suit and everything i mean everything would be on the suit thousands of little cameras on each side of the suit and then on the opposite sides of the cameras thouandss of littl pixel projectors
  14. Nov 2, 2006 #13

    Yes, I can see where I was misleading there Dave. If the system is biologically based, I'd not really consider it a suit any more that one considers skin a suit, but that's just semantics anyhow.

    I wonder if its possible to 'rig' cis-11-retinal to release a photon instead of absorbing one. This in my view would be the key technology to allowing for our 'optical output dodecahedron' scenario.

    What are your thoughts on creating the output photons by running a current across some kind of biological band gap? I'm far from an expert in this area and I seriously doubt the human body could pump enough current though this band gap to emit photons of a visible wavelength, but I'd thought I'd throw the idea out there to more discerning minds than mine.
  15. Nov 2, 2006 #14
    Agreed. I can certainly see your concern about such a large construction. If I could think of a smaller footprint, I would post it, but if we start thinking much smaller we approach the size of individual atoms. Heck, cis-11-retinal is a molecule with less than 100 atoms or very nearly about.

    Just remember that the human retina is only 0.5 mm in width and it has 5 strata chalk full of rods, cones, horizontal cells, biploar cells, amacrine cells and ganglion cells. I have studied it quite deeply and its mind bogglingly complicated with all of its feedback loops, redundant dendrite-axon connections and its syncitial network of gap junctions. Mother nature is just so damn brilliant that you must wonder if we will ever be able to achieve complex systems comparable to hers at such tiny magnitudes. There's really not much room (if any, I don't know) to build things smaller than she does, because we run in QM and the properties of our building blocks start to display rules that we don't understand anymore (hopefully just for the time being).
  16. Nov 2, 2006 #15
    Rumor has it that the US military has developed an invisibility cloak similar in size to those screens used by projectors. It uses the ideas that you expressed and the screen utilizes organic LEDs to create its output pixels. Its basically only good for rendering objects like tanks and artillery pieces invisible from a very narrow angled light of sight, aka. directly in front of the tank. But they are working on crystal technologies that can add tridirectionality to these oLED outputs which promise to increase this LOS invisibility angles up to 45 degrees. Its easy to see that this would be invaluable on the battle field, especially in areas where the enemy must approach you through predetermined channels like a steep mountain pass, or a long bridge, etc.

    Of course the drawback of this system is that its about 70 years too late since radar, IR and Ultrasound detectors would still pick up the screen's signature. But every hurdle is just another challenge to be surmounted so who knows if the military has overcome these as well.
  17. Nov 2, 2006 #16


    User Avatar
    Gold Member

    agreed order.
    although the miltary might want you to think such things. i've seen in a magazine a suit that did that some time ago. similar to this article:
    http://www.wired.com/wired/archive/11.08/pwr_invisible.html" [Broken]
    although its not even close to the complexity of what we're discussing here.

    WHY ORGANIC???-it would be so much harder to re-engineer DNA. we can barley clone.

    i'v read all 4 Artimis fowl books by ecoin colfer. they use a fictional "cam foil" which does exactly what we're trying to impliment into technology here.
    its interesting what the uses they put it through.....

    another comment, why hundereds of tiny cameras insted of having the minimal ammount. if you take a set of cameras (top, bottom, side, side, front, back.) if you make a digital panorama and then impliment this into your technical projection. you really wouldent need all those extra camerasl.
    but all those extra protrudions such as arms and legs do pose some problems. if a camera on an arm was to look at a part of the torso, which is being projected, it would have to inturprit the projection as if it is real quality. in other words, its gotta really look real from any angle. from a mechanical of orgnic point of view.

    and im sure we'll materialistically engineer our way past all those other problems, radar, thermal, etc.....:wink:
    Last edited by a moderator: May 2, 2017
  18. Nov 3, 2006 #17


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    This sounds extremely far-fetched. Since there have been many responses, I will not delete the thread, but am locking it at this point.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook