Slow motion camera that beats light's speed

Printable View

12-14-2011, 03:00 PM
sycld

Slow motion camera that beats light's speed

http://www.dailymail.co.uk/sciencete...oving-air.html

Quote:

The first stunning images of moving bullets appeared around 50 years ago – now a team of snap happy scientists have trumped that achievement by photographing streaks of light.
Scientists from the Massachusetts Institute of Technology (MIT) needed some very specialised equipment to pull off the feat, as light travels around a million times faster than a bullet.
The team used a camera that collected the light beams at a rate of roughly one trillion frames per second - that's fast enough to produce a slow-motion video of a burst of light traveling the length of a one-litre bottle, bouncing off the cap and reflecting back to the bottle’s bottom.
‘We have built a virtual slow motion camera where were can see photons, or light particles, moving through space,’ explained Ramesh Raskar, Associate Professor of the MIT Media Lab.

Ok, well I don't know why he's saying he can see "photons moving through space." He cannot track single photons. But, what he can do is image a light pulse, of course going at the speed of light, as it travels through space.

Imagine seeing a strobe light or camera flash flash once, and being able to see that flash travel from the source outwards. That's what he's able to do. It's pretty amazing.

Of course it still takes the speed of light for light scattered off objects to reach the camera. But otherwise I'm not sure how they do it or quite why it doesn't violate relativity, which I can assure you it does not.

EDIT: I'm stupid. Of course I know why it doesn't violate relativity.
12-14-2011, 05:07 PM
DaiTengu

It's not actually a camera that takes 500 million frames a second. It's a fast camera that photographs a very fast light pulse that happens X amount of nanoseconds apart, then uses a sophisticated software to stitch them all together.

It's still fascinating though.
12-14-2011, 07:27 PM
sycld

Quote:

Originally Posted by DaiTengu

It's not actually a camera that takes 500 million frames a second. It's a fast camera that photographs a very fast light pulse that happens X amount of nanoseconds apart, then uses a sophisticated software to stitch them all together.

It's still fascinating though.

Yeah, it's a streak camera.

Oh, and I just learned that it only captures one horizontal slice of what's being recorded for each frame, and they actually have to repeat the recorded event to construct a full picture. It apparently took about an hour to generate a full movie.

It's still a couple femtoseconds per frame, but not one entire frame recorded all at once :p
12-15-2011, 09:02 AM
Mad Pino Rage

There is a mirror in front of the camera they have to move very slowly to capture the entire scene. Really neat. Didn't know what to expect but I was hoping to actually see what would be a photon.
12-21-2011, 10:44 PM
sycld

Quote:

Originally Posted by Mad Pino Rage

Didn't know what to expect but I was hoping to actually see what would be a photon.

Heh... well... err I guess it would look like a flash of color about the size of the photon's wavelength? I don't think you'll do any better than this. And that's assuming a not-too narrow wave packet, which determines the probability of finding the photon in any particular place in space. If the wave packet is really, really narrow, there's a probability distribution of wavelengths... so I guess a wavelength with high probability would be "chosen" from among these when you measure it.

The length of the flash is probably determined by the spectral resolution of your instrument, i.e. by how well it can measure the energy of the photon. I would think the better the resolution, the shorter the flash of light. The exact energy of the photon you see is determined in the same way the wavelength is, since the wavelength of light in free space is exactly determined by the photon's energy and vice-versa.

In reality, with real instruments, what the photon "looks" like is more determined by the time and space resolution of your instrument than anything else.

But regardless, it would probably just look like a boring, single, very fast flash. But if you performed many measurements of these single flashes that are all prepared the same way, you can reconstruct the wave function of the single photons, which is very interesting. I think, however, you can do this with many photons at once just as well as with photon at a time, since photons like to bunch up together.

There is interest in making single photon with particular wave functions, for experiments where you just want to affect a system by a single photon at a time.