Mechanism behind the generation of tiny bright spots

[johnthekid]

TL;DR

Mechanism behind the generation of tiny bright spots in night vision devices and slit experiments

Watch these clips


and


The flickering bright spots in the second clip which is a night vision device, are the mechanism behind how those bright spots are generated in the night vision device the same as how the individual photon spots are generated for the slit experiment in the first clip?

 

[Ibix]

Sort of. I don't think the image intensifier is sensitive enough to detect individual photons, but it's sensitive to small numbers of them. Their arrival is random. The average arrival rate at a pixel is the image brightness there, but when you're talking about small numbers of photons the random fluctuations in the arrival rate are large enough to be visible as occasional bright and dark spots. The technical term is shot noise.

So the ultimate source of the spots is the quantised nature of light in both cases, but in the Young's slits experiment it is a very dim source so individual photons are counted after diffraction, while the speckle in the image intensifier is a much brighter source that is still dim enough for arrival time statistics to matter.

 

[sophiecentaur]

still dim enough for arrival time statistics to matter.

The arrival times will always matter.
The image intensifier image can be as bright as you like - just turn up the gain
of later amplification stages - but the pattern you see will still fit the diffraction pattern distribution. It will still be a spotty image which, through shaded eyes or a defocussed photo will look correct.

Human vision never evolved to 'see' individual photons but in very low levels of light, it's just marginal. However, photons with higher energy than visible light are detectable (under total darkness conditions) . This Nature article is interesting.

 

[Ibix]

The arrival times will always matter.
The image intensifier image can be as bright as you like - just turn up the gain
of later amplification stages - but the pattern you see will still fit the diffraction pattern distribution.

In Young's slits, the arrival time statistics make no difference to the final pattern, only how long it takes to build up, agreed.

The second video in the OP isn't Young's slits, though. It's just someone wandering around his house with an image intensifier strapped in front of his phone camera. There is a clear image (meaning multiple photons per pixel per second, much brighter than the single photon Young's slits in the first video), and the speckle in that is shot noise. It isn't single photons, but it is due to more or fewer photons than average arriving at one pixel during one readout time.

Photon arrival time statistics wouldn't matter in the second case if the scene were more brightly lit because the scale of shot noise goes with the square root of the number of photons, so becomes less important as you get more photons. That's why you don't see "snow" in daylight, and why you see less noise in the last few seconds of the video when he walks into a brighter room.

 

[johnthekid]

The arrival times will always matter.
The image intensifier image can be as bright as you like - just turn up the gain
of later amplification stages - but the pattern you see will still fit the diffraction pattern distribution. It will still be a spotty image which, through shaded eyes or a defocussed photo will look correct.

Human vision never evolved to 'see' individual photons but in very low levels of light, it's just marginal. However, photons with higher energy than visible light are detectable (under total darkness conditions) . This Nature article is interesting.

The Nature article is indeed interesting, but somehow the images of these individual photons with higher energy are not published. I'm trying to find the additional supplementary materials to see these images of these higher energy individual photons.

 

[johnthekid]

The bright spots are more obvious, although all the bright spots are rather accumulated, in this clip.



The description of the clip, "The reduction of thermionic emission using non-contact radiant cooling of the photocathode. The temperature at the photocathode starts out at 40 degrees Celsius and terminates at 10 degrees Celsius. Thermal electron emissions are a primary source of noise in an image intensifier. There was no source of photo electrons, what your witnessing is the emission of thermal electrons mainly from the photocathode . This was made possible through the cascading of two gen III tubes."

Is it possible to recreate slit experiments without setting up monitors and other stuffs like in the first clip but instead only with minimal requirements for instance an image intensifier then project it into the wall?

 

[johnthekid]

Another clip.



"A demonstration of radioluminescence caused by strontium 90 sample inside a disc impinging on an X-ray paper.

The scintillations are caused by beta particles (a stream of electron) bombarding the fluorescent material. The radioluminescence can just be seen with the human eye after being accustomed to complete darkness and viewed through 20x loupe.

The image intensifier was used to amplify the image so it could be taken with a camera."

 

[sophiecentaur]

In Young's slits, the arrival time statistics make no difference to the final pattern, only how long it takes to build up, agreed.

I was confusing phase / transit time of the waves and the photon. The photon transit time is subject to the Heisenberg uncertainty principle. You may be able to 'know' (record) when the photon arrived but you don't know exactly when it set off. The diffraction pattern is governed by the wave nature and Path Length is used in the integral.

The idea of Signal to Noise Ratio is important because random internally generated noise is always present and ultimately puts a lower limit on the detection of individual genuine 'flashes'.

 

[Andy Resnick]

TL;DR: Mechanism behind the generation of tiny bright spots in night vision devices and slit experiments

The flickering bright spots in the second clip which is a night vision device, are the mechanism behind how those bright spots are generated in the night vision device the same as how the individual photon spots are generated for the slit experiment in the first clip?

No, these are completely different phenomena. The second clip features a lot of electrical signal noise (the overall grainy texture), while the first clip (at 2:17, and later, when using the Hammamtsu system) uses extremely low-noise detectors to detect temporal fluctuations in the electromagnetic field itself. Night-vision goggles (even Gen 5!) are not that sensitive.

 

[Ibix]

Is it possible to recreate slit experiments without setting up monitors and other stuffs like in the first clip but instead only with minimal requirements for instance an image intensifier then project it into the wall?

Do you mean the classical Young's slits, or do you mean the single-photon Young's slits? The classical version is easy with modern technology (I knocked something together out of parts to hand last year). The single-photon version I expect is impossible with off-the-shelf gear. You need a sensor with an extremely small dark current, and that will not come cheap because it isn't something most applications need.

 

[johnthekid]

No, these are completely different phenomena. The second clip features a lot of electrical signal noise (the overall grainy texture), while the first clip (at 2:17, and later, when using the Hammamtsu system) uses extremely low-noise detectors to detect temporal fluctuations in the electromagnetic field itself. Night-vision goggles (even Gen 5!) are not that sensitive.

But are not both involved electrons hitting the fluorescent screen and then produced bright spots on the fluorescent screen?

 

[johnthekid]

Do you mean the classical Young's slits, or do you mean the single-photon Young's slits? The classical version is easy with modern technology (I knocked something together out of parts to hand last year). The single-photon version I expect is impossible with off-the-shelf gear. You need a sensor with an extremely small dark current, and that will not come cheap because it isn't something most applications need.

What are some of the examples of sensors with an extremely small dark current? Image intensifiers? Both Hamamatsu's slit experiment and night vision devices used image intensifiers. The image below shows the Hamamatsu setup of the slit experiment. It is originally in Japanese.

 

[sophiecentaur]

But are not both involved electrons hitting the fluorescent screen and then produced bright spots on the fluorescent screen?

The spots you see are due to the phosphors on the screen being hit by (fast enough) electrons. Those electrons can be due to the wanted photons hitting the first part of the amplification chain. But those electrons need to be accelerated so that they can actually excite a phosphor. To make the sensor sensitive enough to register a photon you have to give it sufficient gain (accelerate it) and this will always register other electrons which are thermally generated in the device. A basic photomultiplier will always be subject to some thermal noise and to use it at maximum sensitivity (lowest density of incident photons) it will also register the internally generated electrons. SO the screen operates as electrons hit it but not all those electrons are wanted. There is no way of knowing when a flash is wanted or just noise.

That's a basic part of information theory; noise is always with us. You can minimise noise by using very low currents and by drastically cooling the sensor. Observing (well, actually sensing) a very distant astronomical object, telescopes basically analyse the outputs of all the sensor elements and decide whether or not the signal they get from an object is significantly bigger (more photons) than the surrounding space. (Which flashes are significant and which ones are not.) Your eyes are 'marginal' in this respect but they had no evolutionary need for that level of sophistication. Nocturnal animals can have better low light performance but they can have problems in bright sunlight.

 

[Ibix]

Both Hamamatsu's slit experiment and night vision devices used image intensifiers.

Yes, and the watch my parents bought me when I was six is a mechanical watch, same as a top of the line Rolex. Slightly different engineering quality and precision, though, and slightly different price.

You need a sensor that has a low enough dark current that 20 photons/second (or whatever) stands out clearly. And you need it to have a high quantum efficiency so that you get a spot every time there's a photon, or photon doublers, beam splitters and coincidence counters. You probably need the whole thing mounted on an optical table, especially if you're coincidence counting. I have no doubt that you can buy this stuff, but you'll have to look at individual device performance and work out what level of performance you can tolerate.

It's been years since I had anything to do with optical experiments, but I can easily imagine spending four or five digits of pounds on this and still not having something good enough to do what you want.

You really need to think about what rate of photons you're going to produce and do a lot of maths and shopping around for equipment. I'd also strongly suggest finding a local optics lab and asking them for advice. There are almost certainly a lot of noise sources I'm not familiar with when you try photon counting.

 

[Ibix]

Also, think about this: if this experiment was in reach of a dedicated hobbyist, do you really believe YouTube would not be knee-deep in videos of people doing it in their garages?

By all means spec out the experiment, but you will need to do a lot of careful choosing of specific components based on their data sheets and a lot of careful maths, and I strongly suspect the price tag will have more digits than you expect. And you need the risk tolerance to swallow that tag and accept you may not make it work.

 

[Andy Resnick]

But are not both involved electrons hitting the fluorescent screen and then produced bright spots on the fluorescent screen?

If I understand you, that is correct. What is different between the two is the mechanism by which mobile electrons are created.

 

[johnthekid]

Before knowing Hamamatsu double slit experiment and night vision devices, I thought we can produce bright spots with cathode ray tubes. But the light generated on the phosphor or fluorescent screen that is located at the end of a cathode ray tube is not bright spots but more like spread out green beam while the image intensifiers used in Hamamatsu double slit experiment and night vision devices (although night vision devices tend to have accumulated bright spots while the Hamamatsu image intensifier have bright spots that gradually build up from fewer to accumulated as the time goes) give multiple bright spots. How to produce bright spots with cathode ray tubes just like with Hamamatsu or night vision image intensifiers, if possible?

 

[sophiecentaur]

I thought we can produce bright spots with cathode ray tubes. But the light generated on the phosphor or fluorescent screen that is located at the end of a cathode ray tube is not bright spots but more like spread out green beam

That's an entirely different piece of equipment. The CRT beam is designed o produce a high intensity spot which consists of a Coulombs worth of charge every second - how many electrons would that be. (Look up the electronic charge (e). The phosphors will be of a different kind with a short afterglow to avoid smearing the motion portrayal.