Question About Filming High Powered Laser Safely

[DanG]

Film camera... lol. What on earth is that???

As for playing with the exposure triangle, yes, perhaps. The issue is that no one really can say what level of irradiance is "safe" for this chip or that chip so there is a lot of guesswork. The display is not live so yes we should be able to fiddle around in post, but this is kind of unique situation where almost nothing can be locked down in terms of this pattern now, this camera move now. Its all very organic and as such it is seeming very difficult to control. Our ability to add logistical elements is unclear so while it is on the table, on camera solutions are preferable so its a less heavy lift. We also definitely cannot re-film the same paths over and over again to layer takes so no dice there.

[swamidog]

Film camera... lol. What on earth is that???

kids these days....

[Vidal Wolf]

Since your budgets unlimited, buy yourself a pile of cameras.
Maybe a good insurance policy for fun.
And while your budgets loose, I can be onsite to give advice for scenarios, scene ideas and general thoughts.
My home airport's Sacramento. =)

[White-Light]

I can think of a possible programming solution but there's nothing that exists currently that I'm aware of.

My thoughts are how about using a Kinect to detect the camera position and then some kind of feedback to the control software to move the BAM position relative to the camera?

It would require module access to alter the BAM on the fly directly from the feedback from the Kinect (ie Pangolin's help in opening this up) and then relative scaling between the BAM width and room width as seen by the Kinect to ensure the camera and BAM stayed in sync. I think it's possible and would probably achieve your goal, but question whether unnecessarily complex?

I've seen a camera used for surfing (called solosho - http://www.youtube.com/watch?v=SE03EesFY9M)) where the surfer attaches a transmitter to his wrist and the camera is on a motorised tripod and tracks the signal from the transmitter to always keep the surfer in shot. Somewhat different in that it only pans whereas in your case the camera is moving. However, it shows camera tracking is possible and we know Kinect is precise, so I'm guessing the big issue here is on the fly BAM adjustment as the tracking is possible.

[AnthonyGarcia]

While I do not have much information for you since I am working in post production (editing, titling, graphics) I love that you are giving an effort to make sure the equipment is safe. On a music video I worked on (a verrrry high budget one at that) the DP, even with the knowledge that lasers can mess up a camera, decided to weave into the beams. Just as we predicted there were camera issues that arose after a few single beams hit dead center.

[logsquared]

Diverging the laser would probably not help unless the beam was made much larger than the aperture of the camera. It may actually be worse for the camera. The spot size on the ccd would be a ratio of input beam diameter and focal length of the lens system.

A polarizer on the camera and linear polarized projector should work. Try it in the shop with a cheap 50mm camera filter and a camcorder.

[mixedgas]

Tests with a 750:1 polarized Hene and a Polaroid based polarizer indicated that dense "Aquafog" did not significantly roll or rotate the polarization of the scattered light at distances of 5 meters. I did not have a small particle hazer available to try with during this test. So the scatter from the smoke dims just like any other "Law of Malus" effect with rotation of the polarizer.

If the fog particles are smaller then one wavelength of light, you'll get isotropic scattering and then the polarization will roll somewhat.

So we're looking at possibly electronic scan fails using a beamsplitter in front of the camera, or placing string pots or MEMS accelerometers to measure the camera position. MEMs 6 DOF gyros with a Kalman filter are cheap. Some scanning laser rangefinders for robotics are down into the 400$ range, so sensing camera position is possible as well.



Dielectric mirrors on the galvo do rotate the polarization somewhat significantly with angle changes and wavelength changes, so one would have to use silver or metallic mirrors.

I'm now going to trademark the terms Nine-Dee Beamshow(tm) and Stereo Beam Effects(tm) using two scan heads or a polarization flipping scan head.. You know, 3D x 3D = 9 Dee... I don't know why you'd want to do it, but you could route scatter from polarized beams to different eyes if the audience is wearing 3D glasses...

Steve

[tribble]

Steve, I love that I can throw a random idea out there and you have the gear and the time to test it almost immediately. :-) That's research, right there. :-)

[DanG]

Thanks for all of the feedback guys! I've got a bunch of stuff to try. I appreciate it!

[White-Light]

I have another suggestion based on what Vidal Wolf said.

I'm came across this dash board / pocket cam whilst looking for a crash cam today - the Mobius Action Cam.

It's very similar in style to the keyring cam that Andy Con bought some time ago, but a real step up in quality.

Anyway my point was, these can be tripod mounted (there's a mounting bracket with a tripod screw socket, so could be fixed to a steady cam or rail system) and the quality is excellent (at least on non laser stuff), but mainly they're cheap at around $70 per cam.

So if money is no object, then you could buy one and if the quality is OK on lasers, use them as disposable cameras as it's only $70 per damaged cam.

Here's a review - stick with it, it's quite long, but it has footage and day / night tests in it further in, and I think you'll be amazed at the quality, especially when it's mounted on a steady mounting as when in a car at 17 mins 45 secs in. (Some of the earlier footage suffers from shake as its hand held / helmet mounted on a bike).

There's also a link on where to buy them from in the Youtube description if you press the Youtube symbnol in the video player bar and go to the Youtube version, and there are links to original samples you can download (without the Youtube compression) to see the actual output quality.

I suggest you watch it at a minimum of 720P.



You'd have to try it on lasers, but it might be worth a look.