A few things to consider: Fast lenses do not always translate into good UV lenses. Even with an aperture of ƒ/0.95, the modern coatings on most lenses filter out a lot of UV. Also, it's not enough to let in lots of light with a fast lens: the lens' optical scheme must allow for focusing in the UV spectrum, which is difficult, as the lens is designed to focus visible light. This explains the "glow" around the flowers in the video. Even older glass lenses aren't that good at transmitting UV - the best UV lenses are made from quartz, which does not block UV, and can focus UV light better than glass. Whilst the Baader-U is one of the best UV passing filters, it is very expensive, and the filter is non-standard size, so a step-up ring is required to attach it to a camera lens. Also, the Baader-U is a dichroic filter, which works by interference, rather than by absorption. The angle of incidence is therefore important, meaning that at wider angles, discolouration occurs at the periphery of the image. Conversely, a ZWB/UG filter stacked with an IR cut filter works by absorption, so the angle of incidence is not a factor. Since UV is not a colour - it's monochrome - removing the bayer filter from the sensor can also help increase UV sensitivity. Of course, you would only do that with a UV-dedicated converted camera.
Achieving good UV results with a glass lens is possible, although you need to find a lens with a low number of elements. Lens coatings are also a problem. Quartz lenses are very expensive, but even with them you are limited by the sensor. The thickness of the sensor itself blocks some UV, and like you mentioned the color bayer layer affects sensitivity too. After everything said above, we're only talking about MWUV to LWUV photography. Getting below 300nm is though. Using an analog ITT with a quartz lens and filters could get you down to the 250nm mark and below. But again, you'll need a special thin sensor designed to capture the UV if you want to go further. KolariVision are selling a good UV pass filter, it is also expensive but saves the trouble a bit from connecting a telescope filter to a regular lens. Veritasium achieved pretty good results though.
Note quite. There is a difference between the near and far infrared spectrum. Those head cameras senses the longer wavelengths of infrared. And that is something you can not pick up on you standard camera and why infrared sensors would be confused of just room temperature objects. Naturally the spectrum is more of a continuum. But anyone that knows anything about black body radiation also knows that even a human eye can see heat... if it hot enough. The warmer the object is the shorter wavelength light it gives off. So yeah. Even a standard camera can see heat. But that is not what people generally mean when we talk about thermal cameras. It is the ability to see a wider spectrum of low wavelength light that is what makes them special (And why they where so expensive in the past. There more affordable today.)
No. Most of the time when referring to Night Vision people are referring to Image Intensifier. That is the light signal is amplified and that is also why you get a lot of noise in those images. It is true that near infrared cameras have been used as night vision cameras to. But in those cases you need to have a infrared spotlight. It is more common when you doing wild life photography or dark room photography. But the military do not use those sort of cameras anymore. They did experiment with it during WWII but the problem is that when you have such a spotlight (which can be quite heavy to have with you to) the enemy can also track you. So it only effective as long as the enemy has not developed there own IR technology. Unlike Image Intensifier. One need to be careful when one use the term night vision because of that. It can mean many different as there is more then one way to achieve night vision. But as the video showed, modern digital cameras are indeed sensitive to both near infra red and near ultra violet but it generally filtered out.
Cythil Night vision cameras use infrared LEDs and sense the reflected radiation. I’ve never seen any (commercial, non-militar) camera using an image intensifier...
So to shoot in the ultraviolet, you need to take the ultraviolet blocker off your camera, (this is the first I've heard of this) and you need an additional filter to block out the infrared. The main things I learned from this video is how little ultraviolet light there is and how much dedication Derek has in making videos.
>you need an additional filter to block out the infrared Not necessarily, no. Other than the UV filter (which is just a coating on the outermost piece of glass you can touch), there is low pass filter, IR filter, and bayer colour array on any piece of digital camera sensor. Derek says he asked to remove both the IR filter and the UV filter so that with the right filter that he could put freely on the front of the lens later, the camera can capture the specific electromagnetic spectrum he desires. If he didn't remove the IR filter, he wouldn't need to buy that expensive IR filter. But if you remove the IR filter but keep the UV, which many photographers do to their old digital cameras, you'd get.. IR camera. Look up IR photography, the images are stunning.
I love doing infrared photography. My old dslr doesn't have it's filter removed however, so I need multiple second exposures even in full daylight. This has an added esthetic benefit for me: it gives a soft ghostly feel to them because everything that moves gets smoothed or becomes ethereal
Never wonder how difficult would be to set up this video, now I do, what a challenge! Congratulations for the perseverance, focus and great work put on that video!
I've played with infrared photography, but never ultraviolet. The same optical issues you discuss. You used to be able to get colour IR slide film from Kodak which shifted everything down a notch with near infrared rendered as red. I thought of the results as "Barbie World", green plants rendered in red and blue (IR -> red, green -> blue), a fuchsia world. The dreamy black and white moonlit look is much more my speed...
Thank you for saving a bunch of money 💵. I have a travel zoom camera that I removed the UV/IR blocking filter. It works great for IR. I thought that since my CMOS sensor is open to full spectrum I would just buy a filter for blocking out visible and IR. There get really pricey and as you mentioned, in UV light it is actually dark outside. Thank you for the terrific content.
Derek, you are the champion of physics for the common TH-cam audience and an excellent film maker. I think that it would be awesome if do a dedicated video about modern cameras, after all you are great film maker. It can be done like this in Veritasium2 , with you chit chatting in front of us.
The spectral response of the camera will also be having a massive effect on the amount of UV you can actually capture. Standard silicon CMOS sensors like those in digital cameras are tuned to be responsive to only visible light. You'll get very little light being picked up by the sensor outside of the visible wavelengths, I'm kind of surprised you can see that much.
no in all digital cameras whether its point and shoot or dslr, they are sensitive to near uv and near infrared they just put in the ir/uv cut filter to block uv and infrare take that out and you have a full spectrum camera.
Hey Derek! A great thing you can shoot with infrared is nebulae! Particularly emission nebulae. I know you need a telescope setup to do it but I'm sure there is plenty of amateur astronomy groups happy to lend a scope.
Great video! I have been trying to make a UV pass filter (for video), with alternate materials. Infrared is no problem using Floppy Disc IR pass filter in combination with a Night Vision camera that actually switches to an uncoated hot mirror, and amplifies the IR signal, and also has IR bulbs. I also created an Infrared and Ultraviolet spotlight to help with lighting in these tests. After a lot of reading I found (and have tested) that Black & white exposed & developed film will block IR (due to Silver) and pass UV. So I made Lithograph filter to pass UV (hopefully) and it does indeed block IR. I have 2 major issues: 1.) the extremely low UV light available. Your video really helped clarify that there is just not enough UV present. I did try aiming my UV spotlight directly at the lens and in the daytime, and saw nothing (now I see it may be from IR contamination), but at night, I could actually see only the pure UV light although it was very faint. Interestingly enough, a green laser seemed to penetrate the filter, but it is possible the laser just illuminated the film itself. 2.) The camera I am using switches to a filter that allows IR (but is it still coated to block most UV?) and also the sensors amplify IR (but do they discriminate UV?). One last idea, is to use an old disposable 35mm point & shoot camera with the clear plastic (full spectrum lens) with a small uv pass, and visible and IR blocking filters taped on...but I am pretty sure the exposure time will not be long enough. Maybe I can just use an IR blocking filter at night (with amplified Night Vision / uncoated hot mirror, and free from visible and much IR light because it is night) to film powerlines, looking for a little arcing. At least I may have a better chance at night (this is what I got from your video). Also, would filming the sun in daylight hours be a sufficient test to see if the UV pass is allowing enough UV through? Very cool either way! Thanks your video was very helpful, especially after searching 100's of pages for these answers for the last few months.
Don’t forget the quantum efficiency is lower at the extremes of the cmos bandwidth which also adds the extra attenuation about 3 stops , normally a wide open chip will work 280nM to 1180nM , when the filter is used for visible light it will generally see 380 though 720nM.
Have you tried messing around with blacklights with the UV camera yet? I'm really curious how much brighter they'd look. In fact, it would be neat to see how other lights stack up too - halogen, fluorescent, LED, etc. Especially at what point along the spectrum a light goes totally dark in UV
Learning about how atmospheric scattering is affected by wave lengths in the electromagnetic spectrum made me curious... Can you see further away with a red filter, or even an IR filter? I was happy to see the answer was... Yes! Red helps add sharpness to very far off shots, and IR takes the clarity even further. Not as drastic of a difference as visible to UV, but still pretty cool.
I purchased Kolari Vision's $199 UV Bandpass. On full spectrum it works as advertised, tough even through an old film lens f1.4. However, when placed front of a Samsung S9 main camera, the photo retains a bit of visible color. That little 3% is INSANE.
I've played with IR filters and photography and this, though it should be very similar, messed with my head! Thanks for the useful information. I assumed, like you, that it would be fairly straightforward to pass only the UV. What you didn't cover was the sensors: Are standard CMOS image sensors sensitive to UV, or do you require even more light getting through to activate the pixels?
CMOS and CCD loose quantum efficiency quite rapidly when going into shorter wavelenghts starting from 400nm. 20-40% absolute QE is usually max what you can get. Specialized blue-shifted sensors can have higher sensitivity but it won't change much if there isn't much UV light available in the first place..
Michael Steeves I work in this research space and CMOS sensors are fully capable of sensing UV (depends on the sensor and if it has things like Bayer filters applied over etc) but the problem is the rest of the spectrum floods the sensor so you have to cut it out like Derek discusses. Try using google scholar to look up smartphones and UV and you will see the work currently being done in this space.
There are also affordable cameras with mono sensors with much better UV sensitivity - used in amateur planetary astrophotography. With a lens they can handle "normal" imaging.
You can see some degree of fluorescence in the trees. The one on the right is dark as it absorbs the UV light. The one on the left, with full spectrum, is shinier than the center trees under only visible light.
The Baader filter actually does pass a bit of infrared--you can see it if the scene contains bright IR sources or if the IR/UV ratio of the scene is unfavorable.
0:23 not Right ! Derek Hot mirror blockes Ir but let pass uv and visible light . The glass lense itself is perfect for blocking uv But for specially Uv reflecting filter called Cold mirror .
It would be interesting to see Corona effect, like in power substations, with your UV camera set, I was looking for some videos, but found only inspection camera footage spotting that there are coronas over the visible image.
I hope I am not thinking about this in a wrong way, but the colors in the UV light film are still perceived through the visible light spectrum by the observer. I mean, there will still be nuances of grey and other colors. We still don't know what the 'color' UV light looks like. Am I wrong?
Well "color" is a subjective experience that we have while certain cones on our retinas get excited by certain wavelengths of light from the visible spectrum. As no cone gets excited by UV light, we don't get an experience of color for UV light. And so footage in the UV spectrum is necessarily conveyed to us in fake colors.
On the other hand, if we did have cones for UV light, it would be a whole new added color, different from what you get when you experience red, green, blue, or any other color you can think of.
That is what intrigues me. I can't imagine another color besides the ones in the visible spectrum. I know how the cones in our retina make us perceive color, but still, UV or infrared colors are hard to imagine without the colors we already know.
There's nothing to really imagine... It's like trying to see in 4th dimensions, it just doesn't work that way for us. If we had an additional cone, our brains would get additional information. However our brain felt like interpreting that type of cone's activation is how we would perceive UV or any other non-trinary color. Try this: explain color to a born-blind person. Now, try this: come up with a 4th color. It's basically the same exercise.
I am familiar with the idea of explaining a color to a blind person. Red is just red. I understand what you are trying to say and I think you are right. It doesn't matter how our eyes see UV light, it matters what our brains do with that information.
Hi Derek. Just wondering why you didn’t choose to use a quartz lens? Was it cost? It allows through more UVB as well, which would increase the amount of UV reaching your sensor (also more scattering of UV but there you are). I am curious because imaging UV is my research field. Great video thanks.
Hello. I hope you do not mind a quick question (i see you are in this field of research). I have a homemade IR pass filter that works amazing (made from Floppy disc) on a Night vision IR camera (so IR sensor boosting, alternate hot mirror etc). I also have a homemade UV pass filter that blocks visible & IR (made from exposed & developed Lithograph film). It does work but I am having problems finding a good UV source. Even my UV spotlight is showing up very weak (but it does pass). Do you think a florescent bulb would be a sufficient source of UV light for this test? I realize I am up against having a camera geared to amplify IR and very low levels of UV light in the environment. I cant figure out if the filter is too opaque, the UV light source is too weak, or the camera itself is working against me....(or all 3 lol). Thanks for your time!
Thank you for doing this. I won’t be shooting in uv but I kept wanting to know how since I know cameras and that it’s not as easy as putting a filter on it lol
Ok I have a wild question, bare with me here. Can the UV filter you used also be used on a UV lamp to allow say a black light to cause things to fluoresce without the purple hue that is normally seen while using a black light. I am essentially trying to create that piece of equipment for a very odd project. Any advice you have would be greatly appreciated
i have a gopro we modified to take other lenses and we removed the uv/ir filter, some glass stovetop look really funky when filmed, even when we filmed visible + ir, it almost looks like cherenkov radiation but purple.
You know that it's easy to record infra red and ultraviolet simultaneously and split them in the editing process after. Your color camera pixels should split it up for you. You just need a filter to get rid of the visible light.
awesome video. I would like to take photos using black light lamp. I prefer don't take off that uv filter from my nikon D7000. So, do you know if exist cheap cameras that don't have this uv filter installed in? Can be analogic cameras. I intend to use nikkor lens.
For better UV sensitivity camera needs further modification - debayering or removal of color filter. But there monochrome Leica camera's which are far more sensitive than color camera's
Would you try out UV-flashlights with that camera? That would be supercool - to illuminate objects with UV and looking that from UV-camera. :). I bought along ago small (thumb size) one and recently regular which was "60W-90W". ( It is closer to 10W) but nevertheless that 10W shows objects like they were from different world. UV is invicible but it triggers different matters to glow in their own way in visible spectrum. Don't look directly to the source and point the light so that you are not accidentally looking direct reflection at close range. I.e surfaces like shiny bathroom tiles may act as UV mirror. If anything doesnt glow, it either absorbs or reflects the UV.
Any idea how to make a uv only scope? Something like a spotting scope versus a camera. I saw plans for one years ago and now that I want one I can’t find anything.
also, there plenty of cheaper filters that transmit uv but block visible and infrared. just do a search for 'shortpass filter' or 'short pass filter' - they are used all the time in fluorescence microscopy.
go to a hauted location and see if you can catch some interesting things. I would certainly try it since it can detect light that we cant with a human eye
It would be quite interesting to know, how well car windshields block UV Light, and if there are any differences between car brands, makes and models. We would learn how important is to wear sunglasses when driving, where you are sometimes forced to look at the sun, sun reflections and sun-glare. While we’re at this, do car windows block uv well enough? Or should we be using sunblock on our arms?
Not sure if you'll see this from a 3 year old comment, but my 2015 Toyota's windshield block probably all uv. Using the Baader U, I need to expose for at least 5 sec to see the 400nm the baader passes, but another filter stack I have, that cuts at about 390nm, doesn't see any thing through the windshield while inside.
I've heard that some people who have had cataract surgery can see UV light afterwards, as the artificial lenses used do not block UV light whereas the natural lens in the eye does. Apparently our eyes are actually more sensitive to UV light than blue, we just don't see it because most of it is blocked by the lens. According to these people, UV blacklights appear bright blue after the surgery!
Very nice explanation, I was interested in the process of filming UV to see if I could easily modify a phone camera to check if I applied correctly the sunscreen on my kids.. apparently it is not so easy😅. As you said that the Atmosphere absorb most of the UV light, did you think about using An UV lamp for you closeup videos?
Can you make a video walking through a city with the uv filter, it’s pretty interesting to see the world in a whole different color spectrum, having this camera you could see what the stress imperfections in glass. I’m wondering what a prince Rupert drop would look like.
Could it be the shallow depth of field of the UV is really more an issue of focus? Since IR and UV are different frequencies than the visual spectrum colors, then the focal length is different for those frequencies. Even the visual spectrum colors don’t have matching focal lengths, so lens technologies and compromises, along with electronic image manipulation or emulsion technologies (ahhh, film) are utilized. Increasing your stop helps the problem, but it’s a compromise. Check out the technology of Technicolor, and how those films shot in the 1950’s were so beautiful and lush. Compare that technology to films shot, say, in the 1970’s when color film with all three color emulsions were captured on one strip of film. The latter doesn’t have the sharpness because now the visual spectrum had to share a common focal length. I enjoyed your video. Thanks!
I like the concept of the World in IR. But then try getting some other, narrower-band filters. Like, one that really only allows IR through at the general spectrum of what the human body gives off.
So does it mean that the Hot Mirror blocks UV too? Does the sensor naturally capture not only IR, but UV as well? I have a camera, which has the ability to night vision with IR. How does it change from IR to visible light? It is very small and does not appear to have any mechanical parts inside. Is it able to capture UV?
It depends on some factors and unfortunately there is no general answer. However, since it is able to record in the near infrared spectrum, I guess that the camera removes the IR filter mechanically when in night vision mode which is a good start since IR filters also tend to block UV. However, you still need to find out what CMOS type is used (frontside or backside illuminated). If it is frontside illuminated, the UV photons will have a harder time getting to the photodiodes of the sensor because they need to pass through the wiring of the sensor first which decreases the UV sensitivity. Furthermore, you also need to check how the spectral response of the CMOS sensor looks like since most Bayer filters (the color filter matrix on the photodiodes) do a very good job at absorbing UV light which is not what we are looking for. It heavily depends on the Bayer sensor though, some still let a bit of UV pass while others nearly fully block it. Another factor is the camera lens itself. Most lenses absorb UV light to some extent and the more glass you have between your sensor and your UV source, the more UV will get absorbed, so a "primitive" lens with less optics inside will probably give you better results. For best results you would need a special fused silica lens but those are usually really expensive (just like proper UV filters which block IR as well). I guess that you just need to give it a try and see how the camera performs.
There's a filter in front of the sensor (typically a Bayer filter) that is a grid of tiny red, green, or blue filters, one color only above each photosite, that only permits light of that color to strike the photosite, which the camera hardware then interprets as a pixel of that color. Even though the UV filter has been removed and there's now a filter that permits only UV light to strike the sensor, the Bayer filter that remains will still pass UV light, and the camera will interpret photons that strike a particular photosite as that "color". There may be intrinsic differences between the pigments used for each of the RGB filters, however, which will bias which wavelengths of UV they can pass, and thus the resulting image isn't interpreted as neutral gray. There are sensors that have no Bayer filter (true multispectrum sensors), and thus would interpret a UV image as a monochrome gray image.
FYI fstop is often unreliable when talking about smaller format cameras since manufacutrers are using the 35mm equivalent value rather than adjusting it to the smaller size. This results in the f number representing (roughly) light per unit area of the sensor rather than total light collected. Basically f 0.95 might not actually collect the same light as an f 0.95 lens for a 35mm camera.
T-stop gives you the actual amount of light that gets through. F-stop number comes from the size of the aperture compared to its distance form the sensor and the length of the lens. F-stop does not accurately represent the amount of light that gets through since it does not take into account how well the glass in the lens transmits light. Really F-stop gives you different depths of field. With a macro lens, the closer you get, the dimmer the lens gets for example. An F2.8 lens could be T4 fairly close up and T8 really up close... Still always giving you the depth of field of F2.8...
@@0ooTheMAXXoo0 fstop is usually a pretty accurate representation of how much light is getting in. You can use t stops if you want to be more precise, but that's beside the point. I'm saying that crop factor is not being applied to the aperture of the smaller format lenses. At f0.95, he's getting the same light per area as an f0.95 lens on a full frame sensor, but only half the total light since his crop factor is 2.
Where can I get UV pass filter? And do you know where to buy cutter to cut these filters (IR & UV filter)? Would like to mod my amateur cameras, Sony A7r III, Sony A7s II and flashes.
What does red hot metal look like with the uv camera ?? Because a good bit of electrons moving down to a lower energy state release photons in the 120nm range not sure if the camera can pick up that small of a UV wavelength curious if u tried it and if it worked or not!
Interesting how the other spectrum looks. I did a bit of IR and that was just 720nm filter on a regular unmodified dslr. Had to do a long exposure in order to capture enough light... Anyway, something else got me really curious though - you're filming outside yet your audio is crisp and clear. Are you on a lav? Or what's your mic setup?
That "10% is UV light" figure is the portion of photons in the UV spectrum compared to the rest of the Spektrum, or is it something else? Like compared to the UV+ VIS + IR spectrum, or some other measurement like the energy portion?
The filter is not for a "two inch telescope". That's not accurate. That sounds like the aperture of the scope is two inches. The two inches are of the focuser barrel of the telescope. I could have a "12 inch telescope" with a 2 inch barrel and that filter would fit.
where's a good place to get adapters to mount a 2" telescope eyepiece filter (Baader-U) to the standard metric filter threads on camera lenses? this is a rabbit hole
Thanks for the great video! Now I've got the Baader-U Filter, but it's got a strange filter diameter. Can you please tell me what adapters are needed to use that filter on usual photography lenses? Thank you very much!
@@stefan_becker so ill need a ring to attach it because i dont have an old lense wich smaller lense size and what did you do with the camera to have the same effect as veritasium ? will you be able to sell your setup camera lense and filters ?
I was wondering if this works in the reverse if you were to use a pass filter. What I'm asking is if removing the hot mirror and putting an optical pass filter in allows you to record in 365 nanometer UV spectrum; could you also put that same optical pass filter in front of a high power LED flashlight and make it a UV flashlight?
A few things to consider:
Fast lenses do not always translate into good UV lenses. Even with an aperture of ƒ/0.95, the modern coatings on most lenses filter out a lot of UV.
Also, it's not enough to let in lots of light with a fast lens: the lens' optical scheme must allow for focusing in the UV spectrum, which is difficult, as the lens is designed to focus visible light. This explains the "glow" around the flowers in the video. Even older glass lenses aren't that good at transmitting UV - the best UV lenses are made from quartz, which does not block UV, and can focus UV light better than glass.
Whilst the Baader-U is one of the best UV passing filters, it is very expensive, and the filter is non-standard size, so a step-up ring is required to attach it to a camera lens.
Also, the Baader-U is a dichroic filter, which works by interference, rather than by absorption. The angle of incidence is therefore important, meaning that at wider angles, discolouration occurs at the periphery of the image. Conversely, a ZWB/UG filter stacked with an IR cut filter works by absorption, so the angle of incidence is not a factor.
Since UV is not a colour - it's monochrome - removing the bayer filter from the sensor can also help increase UV sensitivity. Of course, you would only do that with a UV-dedicated converted camera.
Achieving good UV results with a glass lens is possible, although you need to find a lens with a low number of elements. Lens coatings are also a problem. Quartz lenses are very expensive, but even with them you are limited by the sensor. The thickness of the sensor itself blocks some UV, and like you mentioned the color bayer layer affects sensitivity too.
After everything said above, we're only talking about MWUV to LWUV photography. Getting below 300nm is though. Using an analog ITT with a quartz lens and filters could get you down to the 250nm mark and below. But again, you'll need a special thin sensor designed to capture the UV if you want to go further.
KolariVision are selling a good UV pass filter, it is also expensive but saves the trouble a bit from connecting a telescope filter to a regular lens.
Veritasium achieved pretty good results though.
It would be great to see a video of the infra red spectrum
Dude an infrared camera is the same as a temperature sensing camera.
Note quite. There is a difference between the near and far infrared spectrum. Those head cameras senses the longer wavelengths of infrared. And that is something you can not pick up on you standard camera and why infrared sensors would be confused of just room temperature objects.
Naturally the spectrum is more of a continuum. But anyone that knows anything about black body radiation also knows that even a human eye can see heat... if it hot enough. The warmer the object is the shorter wavelength light it gives off. So yeah. Even a standard camera can see heat. But that is not what people generally mean when we talk about thermal cameras. It is the ability to see a wider spectrum of low wavelength light that is what makes them special (And why they where so expensive in the past. There more affordable today.)
No. Most of the time when referring to Night Vision people are referring to Image Intensifier. That is the light signal is amplified and that is also why you get a lot of noise in those images. It is true that near infrared cameras have been used as night vision cameras to. But in those cases you need to have a infrared spotlight. It is more common when you doing wild life photography or dark room photography. But the military do not use those sort of cameras anymore. They did experiment with it during WWII but the problem is that when you have such a spotlight (which can be quite heavy to have with you to) the enemy can also track you. So it only effective as long as the enemy has not developed there own IR technology. Unlike Image Intensifier.
One need to be careful when one use the term night vision because of that. It can mean many different as there is more then one way to achieve night vision.
But as the video showed, modern digital cameras are indeed sensitive to both near infra red and near ultra violet but it generally filtered out.
Cythil Were? They’re still expensive! Even if they’re cheaper, $400 is not very cheap.
Cythil Night vision cameras use infrared LEDs and sense the reflected radiation. I’ve never seen any (commercial, non-militar) camera using an image intensifier...
So to shoot in the ultraviolet, you need to take the ultraviolet blocker off your camera, (this is the first I've heard of this) and you need an additional filter to block out the infrared. The main things I learned from this video is how little ultraviolet light there is and how much dedication Derek has in making videos.
>you need an additional filter to block out the infrared
Not necessarily, no. Other than the UV filter (which is just a coating on the outermost piece of glass you can touch), there is low pass filter, IR filter, and bayer colour array on any piece of digital camera sensor. Derek says he asked to remove both the IR filter and the UV filter so that with the right filter that he could put freely on the front of the lens later, the camera can capture the specific electromagnetic spectrum he desires. If he didn't remove the IR filter, he wouldn't need to buy that expensive IR filter. But if you remove the IR filter but keep the UV, which many photographers do to their old digital cameras, you'd get.. IR camera. Look up IR photography, the images are stunning.
I love doing infrared photography. My old dslr doesn't have it's filter removed however, so I need multiple second exposures even in full daylight. This has an added esthetic benefit for me: it gives a soft ghostly feel to them because everything that moves gets smoothed or becomes ethereal
Never wonder how difficult would be to set up this video, now I do, what a challenge! Congratulations for the perseverance, focus and great work put on that video!
You can get those filters that pass all wavelength of light at the same stores that sell wifi cables and wifi enhancer spray.
Thanks for the info I might check my local WiFi cable store
can i buy some infinite batterys there as well?
I'm selling infinite batteries in wi-fi stores. Only for 99.99$! Contact me to get one! (Sold only in pairs)
Great, that's almost everything on my list! It looks like I just need blinker fluid for my car now!
@@micahphilson Are you insane? It doesn't exist! But i can sell you one for additional 29.99$
Hey Derek, remember who ELSE has a full spectrum camera? Call me, dude 🤓
This is so awesome to see two channels I admire just chatting in the comments haha!
haha - I'm on it!
I LOVE THIS
This gonna be good
Amazing!! First time saw two great channels talking in comments section!!
I've played with infrared photography, but never ultraviolet. The same optical issues you discuss.
You used to be able to get colour IR slide film from Kodak which shifted everything down a notch with near infrared rendered as red. I thought of the results as "Barbie World", green plants rendered in red and blue (IR -> red, green -> blue), a fuchsia world.
The dreamy black and white moonlit look is much more my speed...
Thank you for saving a bunch of money 💵. I have a travel zoom camera that I removed the UV/IR blocking filter. It works great for IR. I thought that since my CMOS sensor is open to full spectrum I would just buy a filter for blocking out visible and IR. There get really pricey and as you mentioned, in UV light it is actually dark outside. Thank you for the terrific content.
Derek, you are the champion of physics for the common TH-cam audience and an excellent film maker. I think that it would be awesome if do a dedicated video about modern cameras, after all you are great film maker. It can be done like this in Veritasium2 , with you chit chatting in front of us.
I wish I could buy cameras and lenses here in Brazil without having to sell my house.
Yes, please make a video about the world in infrared!
The spectral response of the camera will also be having a massive effect on the amount of UV you can actually capture. Standard silicon CMOS sensors like those in digital cameras are tuned to be responsive to only visible light. You'll get very little light being picked up by the sensor outside of the visible wavelengths, I'm kind of surprised you can see that much.
They must pick up enough of the light to warrant including a UV filter over the sensor, otherwise they wouldn't bother.
no in all digital cameras whether its point and shoot or dslr, they are sensitive to near uv and near infrared they just put in the ir/uv cut filter to block uv and infrare take that out and you have a full spectrum camera.
Hey Derek! A great thing you can shoot with infrared is nebulae! Particularly emission nebulae. I know you need a telescope setup to do it but I'm sure there is plenty of amateur astronomy groups happy to lend a scope.
Great video! I have been trying to make a UV pass filter (for video), with alternate materials. Infrared is no problem using Floppy Disc IR pass filter in combination with a Night Vision camera that actually switches to an uncoated hot mirror, and amplifies the IR signal, and also has IR bulbs. I also created an Infrared and Ultraviolet spotlight to help with lighting in these tests. After a lot of reading I found (and have tested) that Black & white exposed & developed film will block IR (due to Silver) and pass UV. So I made Lithograph filter to pass UV (hopefully) and it does indeed block IR. I have 2 major issues: 1.) the extremely low UV light available. Your video really helped clarify that there is just not enough UV present. I did try aiming my UV spotlight directly at the lens and in the daytime, and saw nothing (now I see it may be from IR contamination), but at night, I could actually see only the pure UV light although it was very faint. Interestingly enough, a green laser seemed to penetrate the filter, but it is possible the laser just illuminated the film itself. 2.) The camera I am using switches to a filter that allows IR (but is it still coated to block most UV?) and also the sensors amplify IR (but do they discriminate UV?). One last idea, is to use an old disposable 35mm point & shoot camera with the clear plastic (full spectrum lens) with a small uv pass, and visible and IR blocking filters taped on...but I am pretty sure the exposure time will not be long enough. Maybe I can just use an IR blocking filter at night (with amplified Night Vision / uncoated hot mirror, and free from visible and much IR light because it is night) to film powerlines, looking for a little arcing. At least I may have a better chance at night (this is what I got from your video). Also, would filming the sun in daylight hours be a sufficient test to see if the UV pass is allowing enough UV through? Very cool either way! Thanks your video was very helpful, especially after searching 100's of pages for these answers for the last few months.
Don’t forget the quantum efficiency is lower at the extremes of the cmos bandwidth which also adds the extra attenuation about 3 stops , normally a wide open chip will work 280nM to 1180nM , when the filter is used for visible light it will generally see 380 though 720nM.
Do a video on why many (but not all) leaves are so reflective in the near infrared (the so-called 'wood effect'). Conifers appear to be an exception.
Oh this is really interesting! I've never heard about this! Would love to hear more!
@@TommoCarroll Hey I watch your videos!! Love them! Keep up the good work..!
Chief DVM woo! Always happy to meet members of the Aspect Herd!!! 🙌 thanks for the kind words!
Chief DVM woo! Always happy to meet members of the Aspect Herd!!! 🙌 thanks for the kind words!
I.m very excited to see how can you shoot on infrared using that camera !
Have you tried messing around with blacklights with the UV camera yet? I'm really curious how much brighter they'd look. In fact, it would be neat to see how other lights stack up too - halogen, fluorescent, LED, etc. Especially at what point along the spectrum a light goes totally dark in UV
Great for diagnosing electrical transmission lines and lighting issues .
Very cool! Thanks for covering this topic.
It was a really interesting video! Didn't actually think I'd be too interested, but wadda you knowww, I was!
Learning about how atmospheric scattering is affected by wave lengths in the electromagnetic spectrum made me curious... Can you see further away with a red filter, or even an IR filter? I was happy to see the answer was... Yes! Red helps add sharpness to very far off shots, and IR takes the clarity even further. Not as drastic of a difference as visible to UV, but still pretty cool.
I purchased Kolari Vision's $199 UV Bandpass. On full spectrum it works as advertised, tough even through an old film lens f1.4. However, when placed front of a Samsung S9 main camera, the photo retains a bit of visible color. That little 3% is INSANE.
Maybe some nightsky photography in full spectrum vs IR vs UV vs visible light ?
(from a rather high altitude, you know why...)
thanks man i always find your videos educational but now you are on my favourite topics, so please more about light
This is great. Thanks for doing this. Not sure if it would be as appealing to someone who wasn't into photography, but I certainly enjoyed it. :)
I've played with IR filters and photography and this, though it should be very similar, messed with my head! Thanks for the useful information. I assumed, like you, that it would be fairly straightforward to pass only the UV.
What you didn't cover was the sensors: Are standard CMOS image sensors sensitive to UV, or do you require even more light getting through to activate the pixels?
That's a good question - I didn't look at sensor response
CMOS and CCD loose quantum efficiency quite rapidly when going into shorter wavelenghts starting from 400nm. 20-40% absolute QE is usually max what you can get. Specialized blue-shifted sensors can have higher sensitivity but it won't change much if there isn't much UV light available in the first place..
Michael Steeves I work in this research space and CMOS sensors are fully capable of sensing UV (depends on the sensor and if it has things like Bayer filters applied over etc) but the problem is the rest of the spectrum floods the sensor so you have to cut it out like Derek discusses. Try using google scholar to look up smartphones and UV and you will see the work currently being done in this space.
There are also affordable cameras with mono sensors with much better UV sensitivity - used in amateur planetary astrophotography. With a lens they can handle "normal" imaging.
You can see some degree of fluorescence in the trees. The one on the right is dark as it absorbs the UV light. The one on the left, with full spectrum, is shinier than the center trees under only visible light.
The Baader filter actually does pass a bit of infrared--you can see it if the scene contains bright IR sources or if the IR/UV ratio of the scene is unfavorable.
Fancy seeing you here, Clark 😊
0:23 not Right ! Derek Hot mirror blockes Ir but let pass uv and visible light .
The glass lense itself is perfect for blocking uv
But for specially Uv reflecting filter called Cold mirror .
How does the CCD respond to UV? in particular how much so with the individual colour channels?
Awesome! I've done some IR photography in the past, it's fascinating stuff
i like how this camera makes invisible rays less abstract😊
That Voightlander lens is
I'm really looking forward to the infrared video.
Nice work, like always, greetings from Brazil
Greetings!
It would be interesting to see Corona effect, like in power substations, with your UV camera set, I was looking for some videos, but found only inspection camera footage spotting that there are coronas over the visible image.
I'm trying to build one to detect coronas. Maybe we can exchange some info
Hope u become more active on yt in the near future. Really enjoy the quality of ur content
I hope I am not thinking about this in a wrong way, but the colors in the UV light film are still perceived through the visible light spectrum by the observer. I mean, there will still be nuances of grey and other colors. We still don't know what the 'color' UV light looks like. Am I wrong?
Well "color" is a subjective experience that we have while certain cones on our retinas get excited by certain wavelengths of light from the visible spectrum. As no cone gets excited by UV light, we don't get an experience of color for UV light. And so footage in the UV spectrum is necessarily conveyed to us in fake colors.
On the other hand, if we did have cones for UV light, it would be a whole new added color, different from what you get when you experience red, green, blue, or any other color you can think of.
That is what intrigues me. I can't imagine another color besides the ones in the visible spectrum. I know how the cones in our retina make us perceive color, but still, UV or infrared colors are hard to imagine without the colors we already know.
There's nothing to really imagine... It's like trying to see in 4th dimensions, it just doesn't work that way for us. If we had an additional cone, our brains would get additional information. However our brain felt like interpreting that type of cone's activation is how we would perceive UV or any other non-trinary color.
Try this: explain color to a born-blind person. Now, try this: come up with a 4th color. It's basically the same exercise.
I am familiar with the idea of explaining a color to a blind person. Red is just red. I understand what you are trying to say and I think you are right. It doesn't matter how our eyes see UV light, it matters what our brains do with that information.
That's awesome. I'd love to see what you end up doing with that full spectrum camera.
Hi Derek. Just wondering why you didn’t choose to use a quartz lens? Was it cost? It allows through more UVB as well, which would increase the amount of UV reaching your sensor (also more scattering of UV but there you are). I am curious because imaging UV is my research field. Great video thanks.
Hello. I hope you do not mind a quick question (i see you are in this field of research). I have a homemade IR pass filter that works amazing (made from Floppy disc) on a Night vision IR camera (so IR sensor boosting, alternate hot mirror etc). I also have a homemade UV pass filter that blocks visible & IR (made from exposed & developed Lithograph film). It does work but I am having problems finding a good UV source. Even my UV spotlight is showing up very weak (but it does pass). Do you think a florescent bulb would be a sufficient source of UV light for this test? I realize I am up against having a camera geared to amplify IR and very low levels of UV light in the environment. I cant figure out if the filter is too opaque, the UV light source is too weak, or the camera itself is working against me....(or all 3 lol). Thanks for your time!
Thank you for doing this. I won’t be shooting in uv but I kept wanting to know how since I know cameras and that it’s not as easy as putting a filter on it lol
Very interesting and informative. Thanks, Dirk.
This was super interesting, thank you for posting it
You should keep an eye on weather sites and try a high uv day. More uv is reaching the ground, even uvc the further we get into the solar minimum.
Maybe, in the future, we'll see a video called 'the world in the infrared'.....
Ok I have a wild question, bare with me here. Can the UV filter you used also be used on a UV lamp to allow say a black light to cause things to fluoresce without the purple hue that is normally seen while using a black light. I am essentially trying to create that piece of equipment for a very odd project. Any advice you have would be greatly appreciated
Great explanation!
Derek has become such a great communicator of ideas/concepts! Are you a massive Veritasium fan?
Yes
i have a gopro we modified to take other lenses and we removed the uv/ir filter, some glass stovetop look really funky when filmed, even when we filmed visible + ir, it almost looks like cherenkov radiation but purple.
You can use your full spectrum camera with LED lights in studio as they does not emit infra, neither UV light
3:30 Ummmm did you also build a KoR Mini Metal Foundry? Was that in another video?
You know that it's easy to record infra red and ultraviolet simultaneously and split them in the editing process after.
Your color camera pixels should split it up for you.
You just need a filter to get rid of the visible light.
UV and IR are not colours - they're monochrome.
awesome video. I would like to take photos using black light lamp. I prefer don't take off that uv filter from my nikon D7000. So, do you know if exist cheap cameras that don't have this uv filter installed in? Can be analogic cameras. I intend to use nikkor lens.
For better UV sensitivity camera needs further modification - debayering or removal of color filter. But there monochrome Leica camera's which are far more sensitive than color camera's
Would you try out UV-flashlights with that camera? That would be supercool - to illuminate objects with UV and looking that from UV-camera. :). I bought along ago small (thumb size) one and recently regular which was "60W-90W". ( It is closer to 10W) but nevertheless that 10W shows objects like they were from different world. UV is invicible but it triggers different matters to glow in their own way in visible spectrum. Don't look directly to the source and point the light so that you are not accidentally looking direct reflection at close range. I.e surfaces like shiny bathroom tiles may act as UV mirror. If anything doesnt glow, it either absorbs or reflects the UV.
Could you put in parallel an infrared camera and an ultraviolet camera to photograph the same landscape, persons, and sky, at the same time?
thank you for sharing this knowledge!
Any idea how to make a uv only scope? Something like a spotting scope versus a camera. I saw plans for one years ago and now that I want one I can’t find anything.
also, there plenty of cheaper filters that transmit uv but block visible and infrared. just do a search for 'shortpass filter' or 'short pass filter' - they are used all the time in fluorescence microscopy.
Do you have any recommendations?
go to a hauted location and see if you can catch some interesting things. I would certainly try it since it can detect light that we cant with a human eye
Could you have used a tanning bed lamp to increase the amount of UV light, or would they not even be bright enough?
It would be quite interesting to know, how well car windshields block UV Light, and if there are any differences between car brands, makes and models. We would learn how important is to wear sunglasses when driving, where you are sometimes forced to look at the sun, sun reflections and sun-glare. While we’re at this, do car windows block uv well enough? Or should we be using sunblock on our arms?
Not sure if you'll see this from a 3 year old comment, but my 2015 Toyota's windshield block probably all uv. Using the Baader U, I need to expose for at least 5 sec to see the 400nm the baader passes, but another filter stack I have, that cuts at about 390nm, doesn't see any thing through the windshield while inside.
What does this look like if you utilize a couple UV flashlights? I imagine it would help quite a bit with exposure.
I've heard that some people who have had cataract surgery can see UV light afterwards, as the artificial lenses used do not block UV light whereas the natural lens in the eye does. Apparently our eyes are actually more sensitive to UV light than blue, we just don't see it because most of it is blocked by the lens. According to these people, UV blacklights appear bright blue after the surgery!
Very nice explanation, I was interested in the process of filming UV to see if I could easily modify a phone camera to check if I applied correctly the sunscreen on my kids.. apparently it is not so easy😅. As you said that the Atmosphere absorb most of the UV light, did you think about using An UV lamp for you closeup videos?
That's a good idea!
Thank you very much for this!
bird plumage typically features full UV spectrum and invisible patterns. can you please investigate how to low-cost shoot UV?
The atmosphere no longer filters all of the uvc. I’ve seen readings of 1.2 mw/cm2, circa 600ft amsl.
Can you make a video walking through a city with the uv filter, it’s pretty interesting to see the world in a whole different color spectrum, having this camera you could see what the stress imperfections in glass. I’m wondering what a prince Rupert drop would look like.
Glass buildings will appear black, in most cases. You won't see the stress imperfections in UV. In order to see those, you need a polarising filter.
Could it be the shallow depth of field of the UV is really more an issue of focus? Since IR and UV are different frequencies than the visual spectrum colors, then the focal length is different for those frequencies. Even the visual spectrum colors don’t have matching focal lengths, so lens technologies and compromises, along with electronic image manipulation or emulsion technologies (ahhh, film) are utilized. Increasing your stop helps the problem, but it’s a compromise. Check out the technology of Technicolor, and how those films shot in the 1950’s were so beautiful and lush. Compare that technology to films shot, say, in the 1970’s when color film with all three color emulsions were captured on one strip of film. The latter doesn’t have the sharpness because now the visual spectrum had to share a common focal length. I enjoyed your video. Thanks!
@2veritasium Try filming at night w/ a UV light source. "Black light", etc...
I like the concept of the World in IR.
But then try getting some other, narrower-band filters.
Like, one that really only allows IR through at the general spectrum of what the human body gives off.
You might like this video about the world in IR, then :) th-cam.com/video/GSPI6NFSZM0/w-d-xo.html
So does it mean that the Hot Mirror blocks UV too? Does the sensor naturally capture not only IR, but UV as well? I have a camera, which has the ability to night vision with IR. How does it change from IR to visible light? It is very small and does not appear to have any mechanical parts inside. Is it able to capture UV?
It depends on some factors and unfortunately there is no general answer. However, since it is able to record in the near infrared spectrum, I guess that the camera removes the IR filter mechanically when in night vision mode which is a good start since IR filters also tend to block UV.
However, you still need to find out what CMOS type is used (frontside or backside illuminated). If it is frontside illuminated, the UV photons will have a harder time getting to the photodiodes of the sensor because they need to pass through the wiring of the sensor first which decreases the UV sensitivity. Furthermore, you also need to check how the spectral response of the CMOS sensor looks like since most Bayer filters (the color filter matrix on the photodiodes) do a very good job at absorbing UV light which is not what we are looking for. It heavily depends on the Bayer sensor though, some still let a bit of UV pass while others nearly fully block it.
Another factor is the camera lens itself. Most lenses absorb UV light to some extent and the more glass you have between your sensor and your UV source, the more UV will get absorbed, so a "primitive" lens with less optics inside will probably give you better results. For best results you would need a special fused silica lens but those are usually really expensive (just like proper UV filters which block IR as well).
I guess that you just need to give it a try and see how the camera performs.
The photoreceptors inside the camera need to actually detect the UV, and then convert it into visible light of some colour. How does that work?
The camera sensor just detects photons, if some of those happen to excite a pixel in the sensor, then you will have some signal from that pixel.
There's a filter in front of the sensor (typically a Bayer filter) that is a grid of tiny red, green, or blue filters, one color only above each photosite, that only permits light of that color to strike the photosite, which the camera hardware then interprets as a pixel of that color. Even though the UV filter has been removed and there's now a filter that permits only UV light to strike the sensor, the Bayer filter that remains will still pass UV light, and the camera will interpret photons that strike a particular photosite as that "color". There may be intrinsic differences between the pigments used for each of the RGB filters, however, which will bias which wavelengths of UV they can pass, and thus the resulting image isn't interpreted as neutral gray. There are sensors that have no Bayer filter (true multispectrum sensors), and thus would interpret a UV image as a monochrome gray image.
That's a lot of hard work...
Thank you for nice information.
Amazing!
Why didn't you stack the zwb1 with an infrared filter instead of going with the expensive filter?
I want to see an ultraviolet video of caucasian faces on a sunny and cloudy day with a baseball cap, bucket hat, and a UV filtering umbrella
FYI fstop is often unreliable when talking about smaller format cameras since manufacutrers are using the 35mm equivalent value rather than adjusting it to the smaller size.
This results in the f number representing (roughly) light per unit area of the sensor rather than total light collected.
Basically f 0.95 might not actually collect the same light as an f 0.95 lens for a 35mm camera.
T-stop gives you the actual amount of light that gets through. F-stop number comes from the size of the aperture compared to its distance form the sensor and the length of the lens. F-stop does not accurately represent the amount of light that gets through since it does not take into account how well the glass in the lens transmits light. Really F-stop gives you different depths of field. With a macro lens, the closer you get, the dimmer the lens gets for example. An F2.8 lens could be T4 fairly close up and T8 really up close... Still always giving you the depth of field of F2.8...
@@0ooTheMAXXoo0 fstop is usually a pretty accurate representation of how much light is getting in. You can use t stops if you want to be more precise, but that's beside the point.
I'm saying that crop factor is not being applied to the aperture of the smaller format lenses. At f0.95, he's getting the same light per area as an f0.95 lens on a full frame sensor, but only half the total light since his crop factor is 2.
Derek, you had a UV lamp. Did you shoot any thing lit with that?
I'm curious to see if different minerals glow like they do in low wave uv black light.
A flourecent almp with the coating removed should be a pretty good light source for UV.
Hi, which lens are you using. Would be useful if you could provide a link with its specs. Thanks!
Best UFO hunting camera!
Where can I get UV pass filter?
And do you know where to buy cutter to cut these filters (IR & UV filter)?
Would like to mod my amateur cameras, Sony A7r III, Sony A7s II and flashes.
What does red hot metal look like with the uv camera ?? Because a good bit of electrons moving down to a lower energy state release photons in the 120nm range not sure if the camera can pick up that small of a UV wavelength curious if u tried it and if it worked or not!
Interesting how the other spectrum looks. I did a bit of IR and that was just 720nm filter on a regular unmodified dslr. Had to do a long exposure in order to capture enough light...
Anyway, something else got me really curious though - you're filming outside yet your audio is crisp and clear. Are you on a lav? Or what's your mic setup?
That "10% is UV light" figure is the portion of photons in the UV spectrum compared to the rest of the Spektrum, or is it something else?
Like compared to the UV+ VIS + IR spectrum, or some other measurement like the energy portion?
what a great video! i love it!
thanks for sharing this great content with us!
really enjoyed watching this
Will this camera work for indoor shooting
The filter is not for a "two inch telescope". That's not accurate. That sounds like the aperture of the scope is two inches. The two inches are of the focuser barrel of the telescope. I could have a "12 inch telescope" with a 2 inch barrel and that filter would fit.
where's a good place to get adapters to mount a 2" telescope eyepiece filter (Baader-U) to the standard metric filter threads on camera lenses? this is a rabbit hole
@Veritasium, did you also remove the bayer filter from the sensor?
Just the information I need. Webcam operation coming up.
Can you shoot reflective UV photography with an analog camera?
Do you know if I could shoot UV light at all with the bandpass filter but without the conversion using a GH5?
If you remove the hot mirror you also have different colors - pink leaves, etc.
Thanks for the great video! Now I've got the Baader-U Filter, but it's got a strange filter diameter. Can you please tell me what adapters are needed to use that filter on usual photography lenses? Thank you very much!
did you find a solution for it ? we just got the baader filter and its smaller than the lense
@@zeplintor I use an old Steinheil lens with even smaller lens size and an adapter also by baader.
@@stefan_becker so ill need a ring to attach it because i dont have an old lense wich smaller lense size and what did you do with the camera to have the same effect as veritasium ?
will you be able to sell your setup camera lense and filters ?
I was wondering if this works in the reverse if you were to use a pass filter. What I'm asking is if removing the hot mirror and putting an optical pass filter in allows you to record in 365 nanometer UV spectrum; could you also put that same optical pass filter in front of a high power LED flashlight and make it a UV flashlight?