原创 Looking for help in recreating an amazing color vision experiment (Part 2)

[Continued from Looking for help in recreating an amazing color vision experiment (Part 1)]

Next we point the spectrometer at another area (again, let's say a pinkish area to keep with our previous example) and we tweak our light sources until the new RGB values being reflected from the pinkish area are the same as the old RGB values from the brownish area.

Initially we are amazed to see that all of the colors on the panel appear to be unchanged. Now we take a piece of white card that covers the entire panel apart from a cut-out that reveals only the original pinkish area ... which magically changes into the brownish color. But when we remove the white card to reveal the entire panel, the pinkish area returns to its original pinkish hue.

How can this be?

In fact, what's happening is that your brain maintains a three-dimensional color-map in which every color is weighted in relation to every other color. Thus, when you can see the whole panel, your brain automatically calculates all of the color relationships and adjusts what you're actually seeing to match what it thinks you should be seeing.

By comparison, if you can see only one shape, then your brain has no other recourse than to assume that this shape's color is determined by the red, blue, and green components that are being reflected from the shape. It's only if you can see a shape's color in the context of all of the other shapes' colors, then your brain does some incredibly nifty signal processing, determines what colors the various shapes should be, and corrects all of the colors before handing the information over to the conscious portion of your mind.

All I can say is that you really have to see this to believe it – speaking of which...

Performing this experiment for ourselves
I would love to recreate this experiment and post it on YouTube so that everyone can see it for themselves. As we see from the discussions above, there are main three elements to this experiment: the three RGB light sources, the panel with the multicolored geometric shapes, and the spectrometer (or whatever we decide to use).

The Lights: I don't think that laying my hands on three stage lights (with individual dimmers) along with three pure color RGB filters will pose a major problem.

The Multicolored Panel: The panel with the multicolored geometric shapes is another issue. If we make it out of board – say 1.83 x 1.83 m – then it's going to be a pain to move around (suppose, for example, that I wanted to replicate this experiment as ESC or DAC next year). Also, if we have say 100 colored geometric areas (squares, rectangles, L-shapes, T-shapes) ... then this is going to cost a fortune in paint, because we would be using only a dribble from each can. But earlier today I had an idea...

In the building in which you find my little office, I share the bay with a company called Out of the Box Exhibits. This is rather cool – they make incredibly low-cost trade show exhibits using cardboard structures clad with a tough canvas material upon which can be painted any design their customers desire. Even better, their graphics expert – Bruce Till – sits in the office next to mine. With Bruce's help, I could easily get a vibrant, multicolored canvas panel designed and printed, so all that remains is...

The Spectrometer (or Equivalent): I remember looking into these a couple of years ago and they were not cheap. But technology has progressed in leaps and bounds, so there are several solutions that spring to mind. One possibility would be to somehow connect a digital camera to a PC running some sort of software application such that you could display what the camera was seeing (like our multicolored panel) in real-time on the PC screen.

Another aspect to the software application would be that it would display a set of cross-hairs on the screen and that you could move these cross-hairs using your mouse or the arrow keys. Wherever the cross-hairs are on the screen, you would see a readout of the corresponding RGB values "under" the cross-hairs.

An even simpler option (in some respects) would be to have a special application running on my iPad, which already has an inbuilt camera. But I know nothing about building iPad apps and I have no clue where to turn...

So now it's over to you. Do you have a better alternative to my camera-PC combo or my creating an iPad app concept? Maybe you know someone who can create iPad apps. If you do have any ideas, please feel free to post a comment .... our operators (well, me, actually) are standing by...