Infrared Training Center

Wednesday, January 26, 2011

Windshield Glass Reflections - How to Remove Them

by Bernie Lyon, Gary Orlove, and Jason Gagnon
"I do a lot of windshield defrost testing at different temps. And I wonder is there any way that can keep from having the camera and myself reflected back into the images."
Glass is about 15% reflective in the 8-12 micrometer waveband. If you are directly facing the windshield, you will inevitably get a reflection of yourself and the camera. I'm sure you have seen this.
One option is to change the angle at which you are observing the windshield, see the image below. If you are at point A, you and the camera will be reflected. If you are at point B, the camera will reflect whatever is above the vehicle, represented by C.
image
If there are hot objects or objects with temperature variations above the vehicle, that might make things worse. They will reflect off of the glass. If possible, you could place a high emissivity piece of material above the windshield so that all reflections off of the glass are uniform. A large piece of cardboard or a blanket might do well.
This way, you will observe only temperature changes, not patterns due to non-uniform reflections.
Another option is to use image subtraction techniques (you will have to have software which supports this, such as FLIR ThermaCAM Researcher).
  1. Take an image before running the defroster.
  2. Then save images as you normally would.
  3. Subtract the first image from the succeeding ones. The resulting image will show only changes in temperature and the reflections will have been eliminated. See the series of images below:

3 comments:

  1. How about reflections on/from shower walls?

    ReplyDelete
  2. The same techniques apply. The most practical is to try to make the reflected temperature uniform over the shower wall. One way to do this would be to employ the shower curtain as the source of reflected energy in the shots.

    ReplyDelete
  3. Many people do a field comparison check of two meters,
    and call them “calibrated” if they give the same reading.
    This isn’t calibration. It’s simply a field check.
    It can show you if there’s a problem,
    but it can’t show you which meter is right.

    Calibration Equipment

    ReplyDelete