Rust-Oleum Fluorescent Green is the color being used on the plywood sides of the green vision tetras. There is a built in function in the camera that automatically attempts to adjust contrast and brightness until it has roughly equalized it (meaning about 50% of the pixels the camera is reading fall on one side of the spectrum, and 50% fall on the other). The good thing about this is that it permits colors to be recognized more accurately and consistently overall but the bad thing is that there aren't too many shades of brightness and contrast between MOE Green and the Rust-Oleum Fluorescent Green. Fortunately I think the RGB is different enough to maybe differentiate us from it, not that I particularly think the RGB has to be different at all, as I will explain later.
Here is the Rust-Oleum green indicated:
http://www.rustoleum.com/Product.asp...ct_id=37&SBL=1
Here is MOE Green: (Just check out the text/pictures)
http://www.moe365.org/
As Dave commented, there are many ways to avoid having your robot track another robot in autonomous. The camera draws the smallest possible square around any patch of the color it's looking for. It then reports the 4 coordinates to you. If you find the area of the square (not hard) you will be able to tell the relative size of the object you're looking at. Because it reports the coordinates instead of just the size of the picture it is also possible to tell the height of the object you're looking at, depending on how you mount your camera. The third way would be to limit the movement of your robot to your half of the field or to the vision tetra positions (harder to implement than to say, but possible). For something else to be the same size, height, color, and in about the same place as the tetra's panel it pretty much has to be intentional, even for we teams who are natural greens.
In regards to strategically painting your robot green. I do not foresee any physical damage coming to your robot from this because no one will be moving that fast if they're using the camera alone to guide them. Do not take this to mean that you can go on an intentionally distracting path in autonomous mode and expect to come out fine. I suspect that many people will not be using this camera at all and will, instead, be running their side of the field around at relatively high speeds in autonomous modes. If you got close enough to a side that your robot was relatively the size of a vision Tetra to an opponent looking with the camera then you'd probably be in the path of someone running around. There would be a big boom and you would feel silly because not only your distraction didn't work, but also your robot got broken before driver mode even started.
So few people are going to end up using the camera, which will required a phenomenal amount of debugging, that building a robot with the intent of distracting the tetra-tracking robots would be an ultimately unsuccessful strategy. If you did not build a robot specifically designed to distract other robots then it would probably be unable to take the abuse of the circuit riders.
Not to mention that intentionally distracting the camera is illegal via R26 and completely against the spirit of gracious professionalism.
PS. All of the information that I posted in this post is included and easily accessible in your Camera manual pdfs. If you want to know more go look it up. They do not list a specific focal length in any of the documentation I've read so far, so it will probably be up to testing. The range probably is dependant on the focus of your camera and the resolution of the camera's image.