Circular Polarisation and Infra-Red Photography

4 min read

Deviation Actions

Okavanga's avatar
For those interested in infra-red photography I have posted two near identical images of a winter scene, taken with my Canon 20D R72 converted camera. In one image I had set a circular polarising filter to allow only polarised light through (Winter Scene IR CP Image)and in the other the filter was set to allow all light through (Winter Scene IR non-CP Image). The shots were taken only a few second apart and early in the morning. The Sun had just risen and was positioned horizontally almost precisely at right angles to the direction of the shots. At this angle, polarisation effects are expected to be greatest. I have very carefully set the same white balance point for both shots, a black point on the metal step of the telephone pole towards the top of the right-hand side. For presentation purposes I have also sharpened each image to the same degree using a high-pass sharpening method in the Gimp. Otherwise no further editing has been carried out. The high ISO of 1600 has given rise to some noise, clearly visible in the sky.

You will notice that the two images are markedly different, with the polarised light image showing a much darker sky and a different tonal balance in the various foliage.Inspection of the RGB channels for both images shows that such differences are in all three channels. There should be no real surprise that the images are different as infra-red light behaves just as visible light and can be present in polarised form. Normally, we expect visibly blue skies to be dark or even black in infra-red photography as infra-red light is not scattered (Rayleigh and Mie scattering)to nearly the same extent as visible light. However, in many situations high level cloud can cause scattering resulting in non-black skies. Using a polarising filter, provided the angle to the Sun is correct, appears to remove this problem.

I am unclear as to the extent of use of polarising filters with infra-red photography, but there would appear to be the potential for further studies in this area.

Winter Scene IR non-CP Image by Okavanga Winter Scene IR CP Image by Okavanga

Edit 10/12/10: I've added the two images here. As a newbie I'm still working these things out. The one on the left is the non-CP image, that on the right is the CP image.

Edit 12/12/10: I've added this spectral response curve for the Canon 40D, available on Canon 40D Spectral Response by Okavanga
The response curves covers the visible and near IR regions.

CSS Journal Coded by FleX177

© 2010 - 2024 Okavanga
Join the community to add your comment. Already a deviant? Log In
JonnySutton's avatar
Great article, what is the effect on exposure times?