In infrared photography special filters are used which block visible light (and ultraviolet light), and only infrared light is recorded. It's really intriguing to be able to capture stuff that we can't see with our own eyes. Especially foliage and grass shine really bright, giving an almost wintery look to the pictures, even when taken in midsummer. Below are some examples which show the difference between infrared photos and regular photos. Some more examples of infrared photography can be found here.
Whereas harsh midday sunlight is often said to be a bad thing for convential color landscape photography, but this is far from true for infrared photography. Overcast light sometimes gives quite dull results where harsh sunny light gives a nice contrast to the photo. The difference can be seen in the examples below, where there was either shadow or sunlight on the land parts.
Shadow vs sunny => [island in the shadow] [island in the sunlight]
The effect of a hot mirror => [with hot mirror] [without hot mirror]
The sensors of digital cameras are sensitive to both visible light and infrared light, but we usually want to take pictures of what we see, and we don't see any infrared light. Therefore, cameras are equipped with so-called hot mirrors, which is a piece of glass in front of the sensor that blocks infrared light (as well as ultraviolet).
Cameras without a hot mirror are called full spectrum cameras and, because the infrared light is not blocked, pictures get a strange pinkish hue, like in the example on the left. This is not an actual picture taken with a full spectrum camera, but an approximation that I made in Photoshop.
If you want to take infrared pictures with a digital camera, there are two options:
• You use your camera as such and put an external infrared filter on your lens, which blocks visible light but is transparent for infrared light. The advantage of this system is that you can use a regular camera, but the disadvantage is that you will need very long shutter speeds (tens of seconds, or maybe even more) because of the hot mirror in your camera. Having an external filter on your lens which blocks visible light, combined with a hot mirror on your sensor which blocks infrared light, means that very little light reaches the sensor, which is the reason why you need really long shutter speeds.
• You convert your camera to a dedicated infrared camera by replacing the hot mirror with a so-called cold mirror, which blocks visible light, but lets infrared light through. The disadvantage is that the camera can only take infrared pictures, but it has several advantages. The camera will take pictures with normal exposure times, you don't need any external filter, and you can freely exchange lenses. A bit of care should be taken with replacing the hot mirror though, because it might be that it also acts as the anti-aliasing filter (more about can be found anti-aliasing filters here).
A remote contol photographed with a normal camera and an IR camera =>[DSLR (1/8 s, f/4, ISO 1600)] [IR DSLR (1/15 s, f/6.3, ISO 100)]
In the first case, you first need to see if your camera is suitable for infrared by performing a very simple test. Just take a remote control, press any button, and take a picture while pressing the butting (and aiming the remote control at your camera of course....). Remote controls work by sending infrared light, so this is a perfect way to test your camera!
To the right are some pictures that I took while testing my cameras. The first photo was taken with my normal camera and there is hardly any light seen coming from the remote control, which means that this is not a very suitable camera for infrared photography. The second picture is from my converted infrared camera, and shows a totally different picture!
So if your camera is sufficiently sensitive to infrared, you can put an external infrared filter on your lens, and take pictures with really long exposure times. But it's not very convenient, both because of the long exposure times, and because it is really difficult to compose with a black filter on your system. So you have to take off the filter, compose and focus, put the filter back on, and take the picture. I did this a couple of times, but it made me feel really frustrated and I quickly stopped taking infrared pictures. So I have converted some of my cameras to dedicated infrared cameras, and that's when the fun starts! No more long exposures and no more filters on and off!
In order to convert your camera, you can either send your camera to specialized companies, or buy the cold mirror and install it yourself. The latter option is cheaper and more fun, but since it is hard to work dust-free in a regular environment, you're bound to get some dust stuck on your sensor. But I found a way to remove the dust pretty easy, which is described below.
Camera - internal filter vs external filter
Internal vs external filter => [DSLR] [DSLR and external IR filter] [IR DSLR]
A thing to keep in mind is that pictures with a dedicated infrared camera differ quite a bit from pictures from a combination of a normal camera and an external infrared filter. This is because an infrared camera will only capture infrared light, whereas a normal camera with an external infrared filter will capture both visible light and infrared light, which will give a picture which is somewhere in between a dedicated infrared camera and a normal camera, probably not too far from a full spectrum camera. (As explained above, the external filter will block most of the visible light and the internal filter will block most of the infrared light. As a result, small amounts of both will reach the sensor, making it more or less a full spectrum camera with a very low sensitivity.)
I have tried to show this principle with the photos on the left where all photos were taken with the exact same type of camera (I have both a normal version and a converted one of this camera), so only the filters differ and there is no difference in the sensors. I converted the photos to black and white to make the differences more apparent.
The first one is a normal photo, the second one a photo with a normal DSLR and an external infrared filter, and the third one is from a dedicated infrared DSLR. It is clear that using an external filter results in a picture somewhere between a normal picture and a real infrared picture.
Camera - internal filter and dust removal
Removing dust bunnies => [original] [dust reference photo] [dust removed]
If you're like me and you insist on converting your camera yourself, then you will have to deal with the pleasure of immortalized dust bunnies on your sensor. Since they will be sandwiched between the sensor and the infrared filter, there is no way to remove them by cleaning. As it did get a bit on my nerves to painstakingly clean my infrared pictures with the dust removal tool, I experimented around to find a way that would make it more easy to remove the dust. It took some time, but now I found a way that seems to work pretty well, and it is a lot faster than removing every single dust bunny.
This method is described in more detail here. The principle behind it is that, once you take a photo, you take a second photo after it with the same settings and a clean detailless background, and this second photo will then serve as a dust reference photo. Afterwards, you substract the dust particles, which are captured in the reference photo, from the original photo, which will give you a clean photo.
As long as the dust does not move around on your sensor, you can reuse the dust reference photos, as long as the aperture and focal length match that of the normal photos. But if you want to be sure, just take a new reference photo. Better safe than sorry!
Not every lens is suitable for infrared photography, and there are lists on the internet with lenses listed as good or bad for infrared photography. Most lenses are simply optimized for visible light, and that gives rise to several phenomona which can occur with infrared photography. Below are some of the ones that I encountered.
The problem with some lenses is that they create so-called hotspots, like in the example on the right where you can see a bright spot in the middle. It changes while changing the focal length, but it is unusable at every setting. Hotspots are most often caused by coatings on both the lenses itself, as well as the coatings on the inside of the lens. Fortunately, many lenses don't show hotspots, or only in a very small amount.
Interestingly, the lenses which show the strongest hotspots in my collection of lenses are actually the more expensive ones, which shows that normal lenses are not optimized for infrared, no matter how expensive.
As if hotspots aren't ugly enough, lenses are also a lot more prone to the weirdest flares with infrared photography, once again because lenses are optimized for visible light. So it's even more important to shield your lens from light sources than with conventional photography. Some examples are below, where the flares can be seen as orange artifacts.
Example of flares
Example of flares
Another unexpected effect was the occurence of coma in several wide-angle lenses. Below to the left is an example of that, where the infrared photo shows a lot of coma, but the one in visible light (taken with the exact same lens) shows the stars as pinpoints. To the right is the upper left part of the same photo, to show the effect more clearly. The effect disappears on stopping down, but for night photography you do not want to stop down, so this lens is not very usable for infrared night photography.
Interestingly, the coma is pointing outwards where usually coma points inwards in visible light. I guess this lens design (and the designs of the other wide-angle lenses showing this effect) has some sort of inadvertent overcompensation in the infrared region. Once again, this is one of my more expensive lenses, so this shows again that normal lenses are not optimized for infrared, no matter how expensive.
Focusing and exposure measurement
When it comes to focusing and exposure measurement, there are a couple of things to remember with infrared photography:
Different ways of focussing => [autofocus] [focus with LiveView]
• Infrared light has a longer wavelength than visible light, and will therefore focus differently than visible light. But the focus sensors in your camera are calibrated for visible light, which means that the infrared light will not focus exactly on the sensor when you use autofocus. It is possible to recalibrate your camera for infrared (or, if you have an older lens, there might even be a focus mark specifically for infrared on the lens itself). How much difference it is depends on the lens itself, and in the case of recalibration, it is done on the camera/lens system, and the new calibration will only work perfect for that combination.
Without calibration, the easiest solution is to use LiveView and focus with that, either manually or automatically. LiveView will display the image as it is recorded by the sensor and if the image is sharp in LiveView then you can be sure that the final image will be sharp as well. On the left is an example of this, where autofocus gives a soft picture, whereas manual focus with LiveView gives a sharp photo.
Unfortunately, if you don't have LiveView on your camera, you have no other option but to shoot several pictures with the lens at different focus distances, and determine on the LCD screen which one is the sharpest. From experience I can say that this is a tedious process and takes away some of the fun with infrared. So I you have the option, go for a camera with LiveView.
• The same goes more or less for exposure measurements, since the exposure sensors are also calibrated for visible light. The result from the exposure meter in your camera is a good estimate, but always check the exposure using the histograms on your LCD screen.
If you take an infrared photo with the camera set to auto white balance, then it will probably look very red like the example below on the left. That's no surprise, since we are shooting infrared, but it doesn't make it very easy to judge your picture. So it is best to make a white balance preset in your camera, by calibrating it on something that is rendered white in infrared, like grass or foliage. If you shoot RAW, it won't have any effect on the end product, but, as said, it's a lot easier to judge the picture with a manual white balance.
There is one problem with infrared RAW files when you use Adobe Camera Raw. The white balance sliders in Adobe Camera Raw can't get low enough to get the same white balance that you obtained on the camera, and your photos will look a lot redder in the RAW editor than they did on your camera. Below on the right you can see how an infrared photo looks in Adobe Camera Raw with the sliders as far to the left as possible. The photo is still reddish and the foliage is not at all white, as I would like to have it.
Fortunately, there is a solution to this problem, which is to make a special camera profile for your infrared camera by using a program called DNG Profile Editor, which can be dowloaded for free. First you save one of your infrared RAW files as a DNG file, after which you open it in DNG Profile Editor. Now you go to the tab Color Matrices on the right, and pull the Temperature slider under White Balance Calibration to the left until your reach the desired white balance (foliage looking white would be a good standard). Now you can export the profile after which the new camera profile can be selected in Adobe Camera Raw under Camera Calibration. Your photo should now look a lot better and you can continue as usual with the RAW editor.
The difference an adapted camera profile makes =>[standard camera profile] [corrected camera profile]
Filters can be used in infrared photography, but, unless you have special filters designed for infrared photography, they are optimized for visible light, which means that they might give different results than you would expect.
ND filters work the same way as with conventional photography, but their intensity might change. In my case, my ND6 filter is only 4.3 stops and my ND10 filter gives about 7 stops in stead of 10.
A gradual ND filter photographed with an IR camera
The same goes of course for gradual ND filters, and in the case of my ND gradual filters, the intensity changes so much that there is no visible effect in infrared!
In the picture to the right I am holding a gradual ND gradual filter, and as you can see, there is no dark part to be seen on the filter. Both the dark and transparent part of the filter are completely transparent for infrared light! That's pretty cool actually, but it also makes this filter useless for infrared photography.
Polarizers can be used, but in my case it gives a very ugly color gradient in the sky, see the two examples below. This might not be a big problem if the image will be converted to black and white, but otherwise it's pretty useless.
But there is hardly any need for using a polarizer if you want more contrast in the sky when shooting infrared. Infrared light is a lot worse at scattering than visible light, which means it will give a dark sky by itself, without the need of a polarizer. The same goes for water surfaces, which are already dark by itself in infrared. In the case of other surface reflections it might be beneficial to use a polarizer, but most of it is not really noteworthy.
Below are two examples of using a polarizer. In the example below to the left you can see that having the polarizer turned the "wrong" way makes the sky look brighter, which might be of use of the sky turns out too dark on the picture.
Example of a polarizer in IR => [no filter] [polarizer] [polarizer rotated 90 degrees]
Example of channel swapping => [original] [channel swapped]
Channel swapping is a technique that is used specifically in infrared photography. It exchanges the information from the red channel with the information from the blue channel, giving blue skies.
Channel swapping is done in Photoshop by opening the channel mixer, selecting the red channel, and adjusting the reds to 0% and the blues to 100%. Then, select the blue channel, and adjust the reds to 100% and the blues to 0%. That's it! It's possible to take different values of course, if that gives better results.