Polarizing filters are mainly used to reduce glare on water or similar surfaces. My question is, are there any filters that do the opposite job? ie that allow the camera to capture more glare or reflections on shiny surfaces?

P.S. Main objective is to increase the effect in reflection photography. Any other advice on increasing the reflection over water surfaces would also be appreciated.

  • @zacharmarz: Yes, you are right. I thought that it was one cycle on a revolution, but it's actually two.
    – Guffa
    Aug 9 '12 at 18:27
  • Note a physically perfect polarizing filter will remove exactly half of light. A real filter removes about just as much (leaking about as much of what it's to block, as blocking of desired polarity) so all photos with polarizing filter need about one stop "brighter" settings.
    – SF.
    Jan 17 '13 at 16:21
  • @SF. How much light a polarizing filter reduces all depends upon the orientation of the filter and the orientation of the light striking it. Your statement assumes that the light striking the polarizer is equally polarized in all directions. That is very rarely (as in pretty much never) the case.
    – Michael C
    Jan 10 at 0:35

The answer will be easy to figure out if you understand a little bit what polarization means.

I don't have a polarizing filter to play with, but I do have a physics degree, so here it goes:

Light reflected by certain types of surfaces (such as glass or water, but not metal) is partially linearly polarized. Light reflected under a certain angle is fully polarized.

Linear polarization means that the electromagnetic wave (light) vibrates in a certain plane only, to put it simply. If you rotate the polarization filter to align with this plane, it lets the polarized light through. If you rotate it to orient it 90 degrees to the plane of polarization, it filters it out fully.

Sunlight will contain light of all polarizations, so a polarization filter will only filter "half of it". Reflected light contains more of light polarized in a plane parallel to the reflecting surface: so if you align the polarization filter perpendicular to the reflecting surface, it will filter out more of the reflected light than light coming from elsewhere. If you orient it parallel to the surface, it will filter out less of the reflected light---the effect you are looking for.

So the short answer is: just rotate the polarization filter and find the orientation which makes the reflection look the brightest! This will accentuate the reflections in the photo instead of suppressing them.

EDIT: Here's an extra idea: you could take two photos, one where you minimize the intensity of the reflections and one where you maximize it. Using these two images, you could make the reflection even stronger by subtracting some of the relfection-less image. It'd take some experimentation with an image processing package to see if it is possible to get it right.

  • 2
    I think its incorrect to say that a polarizer rotated to avoid filtering out light of a given polarization will actually INCREASE the effect. It won't mitigate it, but its also not going to increase it. In other words, rotating the polarizer to not filter out the light being reflected is potentially just as good, but not necessarily better than, not using a polarizer at all.
    – jrista
    Aug 9 '12 at 19:11
  • 14
    @jrista: A filter can't increase light, but the polarising filter will let through 100% of the light with a specific angle, and by average 50% of the light with other angles, so it will increase the reflected light relative to the non-reflected light. It works as a neutral density filter for non-reflected light.
    – Guffa
    Aug 9 '12 at 21:14
  • @Guffa: Again, I would dispute that anything is increasing at all. Something, in this case the other half of light, is being reduced as it doesn't fully match the polarization angle of the filter...but the reflections are not being increased. From practical experience, angling a polarizer to the primary angle of reflected light still tends to reduce the overall amount of reflection, as not every single ray of reflected light always has exactly the same polarity...especially off of water. Your almost always going to reduce reflections a little.
    – jrista
    Aug 9 '12 at 21:31
  • 10
    @jrista Everything in photography is relative. Theoretically speaking, not filtering out the reflections whilst filtering out 50% non reflected light will increase the relative strength of the reflections. In practice results will vary as the reflections are not entirely polarized the unreflected light may be partially polarized by the atmosphere.
    – Matt Grum
    Aug 10 '12 at 12:23
  • 1
    Let us take this from a different angle: If you want to objectively INCREASE reflections, you set your camera one stop higher and then you increase ALL of the image, highlights, mattes, shadows, EVERYTHING - including reflections. With a polarizing filter you still have the camera take the photo one step higher(!!!) but then all non-reflections are dimmed. With the filter you first increase the reflections (and everything else) by taking a photo one step higher, then dim everything else than the reflections back down using the filter.
    – SF.
    Jan 17 '13 at 16:29

Any Other advice on increasing the reflection over water surfaces.

Specular reflections like you get from water are stronger the lower the angle of incidence. This means when the light is coming almost parallel to the surface and striking a glancing blow.

This is easy to achieve under controlled lighting. In natural light this means waiting until the sun is low in the sky like at sunrise or sunset.


Yes, you use a polarising filter for that too.

At a specific angle the polarising filter will let through 0% of the reflected light, and 50% of the rest of the light.

Turned 90 degrees the filter will let through 100% of the reflected light, and 50% of the rest of the light.

  • 1
    You mean ninety degrees, not 180 degrees.
    – Staale S
    Aug 9 '12 at 15:32
  • 2
    @Baczek: Why wouldn't it? The only difference is that the circular filter has a de-polariser behind the polariser filter.
    – Guffa
    Aug 9 '12 at 21:11
  • 1
    @Baczek: You are wrong. Circular filters only randomize the incoming (polarized) light after it has already passed through a perfectly standard linear polarizing layer at the front of the filter. This has no visible effect but ensures that the light exiting the filter is for all practical purposes randomly polarized so that the autofocus mechanism in the camera can do its thing. The linear polarizing layer at front still does what it always did.
    – Staale S
    Aug 10 '12 at 10:30
  • 1
    @Baczek: A circular polarizer filters light according to its linear polarization (like a standard polarizer), then converts this polarization into a circular polarization (not a random polarization as Staale S says). Thus, you really get circular polarized light out of the filter. Aug 12 '12 at 12:26
  • 1
    @SF.: Both linear and circular is used in an illogical way for polarising filters, as the terms doesn't mean at all what they do in other circumstances. Most polarising filters are actually both linear (i.e. having an output proportional to the input), and circular (i.e. in the shape of a circle). A linear polarising filter should just be called a polarising filter, and a circular polarising filter should be called something like a de-polarised polarising filter, but correct naming doesn't help you to understand what they do unless you understand the advanced optics involved.
    – Guffa
    Jan 17 '13 at 17:04

Not the answer you're looking for? Browse other questions tagged or ask your own question.