Monday, May 12, 2008

Cameras and infrared. 2 – Digital.

First of all, the only serious problem I have with digital cameras is the ease with which their photos can be manipulated. Fakery is far too easy with these cameras. Using a digital image as ‘proof’ is going to be very difficult indeed. They do have some very useful aspects though, and that does not include the 'orbs' so championed by those who should know better.

Digital cameras see into the infrared. The easy way to prove this to yourself is to take a TV remote, the kind with a red window or visible LED, point it at your face and press a button. You’ll see nothing.

Now, aim it at a digital camera, watch the screen and press a button. You’ll see the remote light up. It’s infrared, and the camera can see it.

You’re probably wondering, then, why you can see it through the camera if it’s infrared, and your eyes can’t see infrared? The screen you’re looking at can’t display infrared either.

The camera’s sensor reacts to light. Visible light, infrared, ultraviolet, all of it. It sees the infrared as light, which activates pixels, which send the information ‘I have seen the light’ to the processor, which then transfers that image of light to the right bit of the screen. The processor simply forms an image based on what the pixels tell it, and since the screen works only in visible light images, it translates the infrared image sent by the sensor into a visible-light equivalent on the screen.

This works best on a monochrome setting because there’s no ‘colour’ in infrared or ultraviolet. They are outside our visible range so we can’t assign them ‘red’ and ‘blue’, they just show up as ‘light’. That’s why it works best as a greyscale image.

Because of this sensitivity to infrared, the optics in digital cameras include a filter to block most of it. It doesn’t block all of it—and the degree of infrared-blocking varies immensely between cameras—but it blocks enough so the infrared image doesn’t interfere with the visible-light image. There are two ways around this.

One is to fit an infrared filter, which looks totally opaque because it blocks all visible light. Only infrared gets though. Set the camera to monochrome (black and white) and look through the filter. You’ll need a camera capable of high ASA (ISO) settings and/or able to operate at slow shutter speeds. You’re also best to have a tripod for this, even though you’ll be shooting in daylight.

At night, where there’s little or no visible light around, you don’t even need the filter. You will probably still need the tripod. I can get away with handheld using a Sony DSC-H5 but even so, most shots are blurred. A tripod is best.

This scores over the use of infrared film in several important respects.

-You don’t have to pay for film so even if nine out of ten shots are rubbish, it’s cost nothing.

-There’s no film to keep cool. Just slot in the filter and you’re in business.

-You don’t have to use up a whole series of shots in one session. You can alternate infrared and normal shots, which means you can take photos of the same scene in both normal and infrared within seconds of each other.

-Best of all, when you look at the camera’s viewscreen you can see what the camera sees. That’s not possible with film. The exception is night photography where you might have trouble seeing the screen. Even so, that’s no worse than the film option.


The downside, as always, is that it’s easy to produce a fake so whatever you find will be hard to use as proof. All the same, you have a good chance of capturing images that you might miss with film, not least because you can shoot away without worrying about film costs. Just remember these cameras have no ‘manual’ setting. If the battery runs out, the camera won’t work at all so always have spares.

Infrared photos taken with these cameras can be grainy, because of the need for a high ASA setting. The DSC-H5 will go to 1000ASA but it’s seriously grainy at that setting. It does let you capture a scene you wouldn’t have been able to get otherwise, even if it isn’t the best photo in the world.

That’s another thing about digital. If you load your camera with 400ASA film, then you’re stuck with that until the end of the roll. A digital camera can be set to 100ASA for one shot, then to 800ASA for the next. It’s far less restrictive.

So what kind of images can you expect? Well, here are some shots of an old shed beside the railway line, close to that old cottage I’ve been looking at.

This is it in daylight, in colour. It's just a shed:

Daylight, monochrome. It’s easier to compare with the infrared images this way.


Here it is again in daylight with an infrared filter. These photos were taken at different times (well, duh!) and from different angles, but they are used here as examples and not a scientific study of a shed:
Vegetation reflects infrared so appears bright. The blue sky isn't emitting much infrared so it looks dark. If ghosts are related to infrared radiation in any way, they would be expected to show up in such images. Not necessarily as perfectly-formed images of people, but at least as recognisable outlines, preferably of the right height and shape.

And again at night, without the filter – first, distant and enlarged:

It's very grainy because there was almost no visible light available. What the camera is picking up here is infrared, and there wasn't too much of that available either. The 'streak' in the top left corner is a twig. There was a nearby bush in shot.

Night, no filter, closer than before and therefore less grainy:

The one good thing about grainy images is that they're harder to add fakes to. Matching the 'grain' would take considerable skill and a lot of time, and fakers aren't likely to go that far. You could enlarge these and look at the overall grain and work out whether I'd added something. I haven't. What you see are the photos I took. Feel free to enlarge them and check.

Infrared images are never going to be perfect. The blocking filter in the camera's optics is a sticking point. I did read an article about how to take those out, and even took apart a couple of old video cameras to try it. I discovered that my skills did not extend to reassembling those optics. It's not a good idea to try that with your best cameras - the old ones I used for that experiment are now scrap.

All the same, if you want to try infrared, the cheapest and easiest place to start is with a digital camera and an infrared filter. Results vary between cameras, so you'll want to try out a few before spending a lot of cash. Get the filter first and ask your friends to loan you their cameras for a moment.

Oh, and make sure you buy one that a) has a monochrome setting and b) has the ability to accept filters. Not all of them do this!

It works with video cameras too.

3 comments:

astrologymemphis.blogspot.com said...

Have I seen the third picture down before? I think I'll take a closer look at it. Also, I was looking again at the second one down just the other day, and meant to ask you: is there someone standing between the right edge of the building and the fence post?

astrologymemphis.blogspot.com said...

I just noticed. The third one down is a different shed, isn't it?

Romulus Crowe said...

Same shed, different angles, lighting and one with the infrared filter.

I notice Blogger won't enlarge these. Sometimes it works, sometimes not. So I've set up an account on MySpace where I can load the pictures:

http://www.myspace.com/romuluscrowe

Unfortunately you have to join to see the pics. Fortunately it's free. Some people have produced remarkably complex sites there but it'll take me a while to work out how to do that.

opinions powered by SendLove.to