Dec 10, 2013

Shedding Light On How To Clear Up High Key Flare

Earlier this week a photographer that I have been training had her very first client shoot. Teary eyed I offered to come along for moral support and to watch my little birdie take her first step from the nest. She did great! Though she ran into a single very specific problem.

All the images on the back of her LCD looked washed out. She was having problems with the high key backdrop causing flare in her lens. I offered few quiet pieces of advice and she was able to solve the problem but the situation had me thinking that this may be a challenge that many photographers face.

So I spent a few hours in my studio recreating the problem exactly as she faced it so that I could show you how to solve it. Every example image in this tutorial was shot at F5.6 and are un-edited in Photoshop. I opted to shoot a still object for the ease of instruction, however, this problem is something any photographer can run into when making any high key image.

The Problem

The light setup is pretty simple; two background lights on either side of the seamless at about camera level pointing towards the center of the background. I also included another strip bank to camera right to light the front of the bottle and create a highlight.

As you can see, even though the bottle is fairly well exposed having no blown out highlights or shadows the contrast is extremely poor do to flaring that is caused by light bouncing off the backdrop.

Step 1: Minimize Background Exposure By Lowering Light Power

The first step to solving this problem is likely also the most obvious. The background lights are simply too bright. You don’t need to blast enough power at the backdrop to put a spotlight on the moon. By simply scaling back the background strobes by a couple stops you can see that the flare noticeably diminishes.

You also might notice that the background is no longer fully white as there is now a subtle gradient. The lights are merely not bright enough to evenly expose the backdrop to create the high key effect. Furthermore, there is still an annoying amount of flare.

Step 2: Angle Those Strobes

Lens flare is caused by light bouncing into your lens and scattering at unwanted angles. In the case of this image the flare is being caused by light bouncing off the white backdrop and directly back at the camera.

By lowering the background lights from being at camera level to being on the floor we are able to angle them up so that the majority of the light skims across the seamless and bounces into the ceiling. Not only does this help reduce the flare even further but it also leads to more even distribution of light throughout the seamless eliminating the gradient problem from the previous image.

Step 3: Introduce Flags

For the most part we now have the background problem nicely handled, however, the image still has a small hint of flare. This remaining flare is actually caused by direct line of sight to the strobes. Even though reflectors are being used to direct light away from camera there is still noticeable spillage.

In order to solve this problem we can set up a pair of flags to block direct line of site from camera to the background lights. In the case of this example image I put up two large 8’x4′ v-flats to ensure that even any oddly bouncing light would be prevented from finding its way into the barrel of my lens.

We are rewarded with a nice, rich, flare free, image.

Why Fixing It In Post Is NOT A Good Idea

Flaring manifests as poor contrast. Lightroom has a nifty contrast slider. It may seem too obvious to simply not worry about flaring in camera and to let the computer solve those annoying flare problems.

This is, however, not good. As you can see in the example image that by cranking up the contrast post exposure we are able to eliminate the flare at a cost. Not only is there a loss of detail due to shadows being blown out but also a blue tint is introduced which ultimately will lead to color problems with the rest of the image if we were to use selective color to attempt a fix.

Hover over the image to see before.

Why A Lens Hood Doesn’t Really Help

Normally, we use lens hoods to control flaring. A lens hood is designed to block directional light that is finding it’s way into the lens from outside of frame. However, in the case of high key flare the haze is being created by a light source that is in your frame and thus you cannot depend on a hood to block the errant light.

Why Gear Really Does Sometimes Matter

Its no secret that while even though we photographers tend to love our gear we also love to remind ourselves that great photography is about the photographer and not the gear. Which for the most part is true but at the same time gear can also often help you deliver higher quality images and allow for more radical situations that would have been otherwise very difficult.

To create the following photo I reset the lighting back to the original set up with the strobes fairly bright at camera level. The scenario that was causing a ton of flare.

I then switched away from the cheaper, entry level, 50mm prime that my student had been using and instead popped a much more expensive 100mm macro lens onto my camera that features fancy coatings while also enjoying higher quality glass.

As you can no doubt see, even with the lights turned up so high that they were causing unusable flare on the previous lens the better quality lens is able to cut through the flare without a problem and is actually creating a higher contrast image than even the best version taken with the cheap fifty.


Creative problem solving is definitely handy when dealing with flare. Even with the cheapest of entry level lenses solid technique can eliminate that flare and allow you to create great images. However, having to constantly tinker with flags and light angles to protect against background bounce quickly becomes tedious, especially when working with a tight timeline and thus it may eventually be worthwhile to explore upgrading to a better lens which can handle background flare much more effectively.


user image You
(will not be published)

  • user image
    Rik Koenig

    Good article.

    One thing’s bothering me, though. In the lens test at the end, you switched from a 50mm to a 100mm. That seems like it would change too many variables to be a controlled test. By changing lens, you are changing angle of acceptance, distance to background, and amount of visible background. I don’t know how much (or even if) these factors would affect flare, but it seems like a better, more controlled test of lens affect on flare would be to swap the cheap 50 for a better 50.

    • user image
      Ryan Cooper

      I agree, and fair enough. Only problem is that I didn’t have a better 50 handy hahaha. Been waiting to see if Sigma comes out with a new Art 50 before investing in Nikon’s new 58mm F1.4. I can’t stand their current 50mm lenses so have been making do without since my last one bit the dust.

      Regardless, though, it was meant as less of a controlled test and more of just an example of how using a more suitable (expensive) lens, in this case, can eliminate much of the problem from the get go. A big part of that suitability is focal length.

      • user image
        MeasMo Photo/Design (@MeasMoPhoto)

        I agree with Rik, switching from the 50MM to a 100MM lens bothers me too because it becomes like comparing apples to oranges and (in my opinion) could have nothing to do with the quality of the lens.

        As you’re probably aware (without going into detail) the proper way to get a white background is generally to create separation between your subject and the background so the light from your background is not spilling onto (or past) your subject. Using this approach you can obtain a solid white background with approximately a stop difference (give or take depending on the specific variables) between your background lights and your key light with no flare (once again depending on specific variables).

        By switching to a 100MM lens you would have obviously had to move your camera further from the background so assuming you didn’t move any of your lights this alone may or may not have solved your problem because of the new distance and angle from your subject and the light sources. Another question is did you also move the subject further from the background. If you moved the camera and the subject further from the background then the lens quality probably played little or no part in the final results and instead you would have probably gotten the same results (as far as lens flare) with a cheaper lens as well.

        This is a great article and I really enjoyed reading it. It gives us food for thought and I guess the only way to be certain is to set it up and test it. Thanks for sharing.

        • user image
          MeasMo Photo/Design (@MeasMoPhoto)

          I apologize for the image I thought it was for the avatar.

    • user image
      Steven Kersting

      Changing the FL doesn’t change the distance to the BG. It does change the proportional distance if the working distance is increased to maintain the FOV @ the subject as opposed to cropping the 50mm shot for the same final composition.

      But the FOV of the lens will certainly affect succeptability.

  • user image
    Steven Kersting

    Minor comment…

    Feathering the lights at an angle does not exactly reduce the amount of light coming back to the lens… For the BG to be white it requires the same amount of light into the lens regardless of the lighting direction.

    What it does do is allow more even lighting as you mentioned. What often happens is the BG lights are too small in spread and this requires that they run at a higher power in order to blow out the edges of the BG… The center of the light spread is “extra blown” (but the camera can’t record that) and it’s that “extra” that causes the flare (and often light on the back of the subject).

    There’s nothing wrong with angling the lights but it can have some negative side affects. The lights have to run harder due to so much light not being reflected back to the camera, and all of that light can add to the spill/ambient in the studio; particularly if the space is small. That can make controlling ratios/directions more difficult.

    A larger/ more even light source for the BG would also have reduced the issue; perhaps adding/changing a modifier. And if that’s an option it is probably the better choice.

  • user image
    Jeroen van Rooij

    Again, a very nicely written and helpfull article for those out there (like myself) trying to get the best out of the equipement they can afford. I’ll stick this one in my virtual toolbox 😉

  • user image
    Juan Raphael Prieto

    I had this diffused edege flair happen when I was shooting a subject naturally backlit from a window into a fairly dark space. The image was totally washed out. Even in the center of the subject! I changed my appreture to see if it made a diffrence but there was no change. I never had that kind of flair happen before and will be looking into solutions if any. Your article came in at a good time as I have been puzzled by how much the backlighting affected the image.

  • user image
    Karen Hoinke

    This was an excellant tutorial…Thank you. I am looking for a way to photograph a wine glass with half filled with wine on a white background….. I changed the white background to pure white…. but didn’t know how to change the background through the glass of wine to look remotely the same color… looks weird. Hope this make sense. Is their a lesson on that? thank you Aaron, this was helpful….. I will try again… :0)

  • user image

    I don’t shoot much with flash but I would add that it’s often not the glass on the lens but maybe the UV filter in front of the lens. At the time I used my EF 50 f1.8 I used it without a protective filter especially indoors because it should be better quality and the lens is really cheap.
    For my more expensive lenses I use the clear filter from B+W and I can highly recommend it (money well spent). I have the feeling that it produces less flare and better contrast in tricky situations.
    Further I think it might be a good idea to move the photographed object further away from the background so the interaction between light bouncing off the background diminishes (by the rule of squares?)

  • user image

    Hello and hope you are fine
    A couple of weeks ago a friend of mine took a picture, I am now attaching photo, need to erase len flares and haze…is there any tutorial from you about this problems ?
    Hope you have a few words with me
    be well

  • user image


    Could this be a demonstration of the difference in direct reflection(mirrorlike / specular) and diffuse reflection? By using a longer focal length you might have excluded all the angles that would give a direct (mirrorlike) reflection. And with that you only have to deal with a diffuse reflection.

    It’s like holding a mirror in your hand. If you hold a small mirror you wil not be able to see as much of the environment behind you (including bright strobes) as when you are holding a very large mirror. I imagine that changing the focal length has a similar (not exactly the same mechanism) effect on the field of view of the reflection.

    Feathering the lights would change the angle of the direct light beams and their reflections therefore moving them out of the family of angles that your lens can “see”.

    Anyway, I know that using longer focal lengths is a trick to reduce reflections when shooting through glass. I think it is possible that a cheap 100mm gives the same performance (flarewise) as an expensive 100mm. A 200mm might even do better.

    Any thoughts? Have you done new experiments since 2013?

    By the way, I love what you are doing here at phlearn!

    Kind regards,