HDR Increments


Today’s Question: Is there a benefit to using one-stop increments for the individual exposures of a high dynamic range image? I have been using this approach, but a fellow photographer recently told me that two-stop increments are just as good. What’s your opinion?

Tim’s Quick Answer: As far as I’m concerned there is no real benefit to capturing photos with one-stop exposure increments compared to two-stop increments when creating a high dynamic range (HDR) image.

More Detail: As long as you cover the full tonal range of the scene, with reasonable overlap between exposures, you will be able to achieve good results for the final HDR image. I almost always separate my individual exposures by two stops when capturing a sequence of images for an HDR image.

The only time I use a one-stop increment between captures is when the images are being captured with the camera mounted on a tripod, I have plenty of time to work without any concern of the scene changing, and I’m feeling especially detail-oriented.

When I have captured HDR sequences with a one-stop increment I have actually perform tests where I create two versions of the HDR image. The first image uses all of the images captured at one-stop increments, and the second image uses every other capture, resulting in two-stop increments. From a quality perspective the results have always been the same with both approaches.

What I consider to be more important than one- versus two-stop exposure increments is to cover the full range of tonal values in the scene. I also make a point of ensuring that the brightest exposure is captured so that the darkest areas of the scene are relatively bright, helping to ensure high detail and minimum noise in those areas.

For more information about my recommended approach to capturing high dynamic range (HDR) scenes, see the article “Optimal HDR Exposure” in the September 2012 issue of Pixology magazine. If you’re not a subscriber already, you can get more details (and free back issues) at http://www.pixologymag.com.