Mike's Web Page
Home ] Up ] Photo Gallery ] Mandelbrot Images ] About Me ] What's New ] Links ]



Infrared (IR) photography is kind of cool. I like the fact that we can make sensors to detect things our eyes can't see. There's a whole ocean of different photons out there.

For those who would like a bit of background on what infrared light is and how IR photography works, click here. I'm certainly no expert on this, but I think my physics background qualifies me to describe some of the elements of it anyway.

I have a couple of different ways to take IR pictures.

First, and simplest, is to attach a so-called "IR filter" to the front of my Canon S2 using the lens adaptor tube. I say "so-called" because these are really visible-light filters that allow infrared light to pass. I have a 58 mm  Hoya R72 filter, which basically blocks light with a wavelength shorter than 720 nm. This attaches via a little plastic lens adapter tube to the front of the S2. This method is easy, and it works surprisingly well. It is surprising because, like any digital camera, this one has an infrared-blocking filter right in front of the sensor. The filter is there because, unlike film, digital image sensors are quite sensitive to IR light, so this has to be filtered out to prevent weird color casts from forming as the IR light hits the red, green, and blue sensors and alters the color balance. The filters are usually pretty good, so it is surprising that enough IR light gets through to make an image. But, if the exposure time is long enough, it will work and yields decent IR images.

The other alternative is to remove the IR-blocking filter from a digital camera. This requires opening up the camera, risking being unable to put it back together again. Fortunately, we were finished with the Fuji FinePix, so I opened it up and took out its IR filter. This works pretty well. I have to hold the IR filter in front of it by hand, since it's just a point-and-shoot model with no way to mount a filter by screwing it on or anything like that, so it's a bit cumbersome.

Either way, the images are recorded on an RGB sensor, so, of course, they have to be converted to grayscale afterwards using Photoshop or Corel or whatever.