Tuesday, September 27, 2016

To HRD, or not HDR? That's Not The Question

It's a Matter of Style

HDR photography has been around for several years.  There are some views that HDR means over processed photos that don't look "real" or "natural".  I don't intend to write a technical blog and there are tons of articles out there about the more technical aspects of HDR and the pros and cons so I won't go into all that here.

In case you don't know, HDR stands for "High Dynamic Range".  Several newer model cameras have an "HDR" setting.  I can't say how each of these camera manufacturers achieve what they produce when you take an "HDR" picture.  What I want to talk about here is the traditional view of HDR photography.

So what exactly is HDR photography in the traditional sense?  An HDR image is usually made by software in post-processing by combining three images of the same exact scene, each at a different exposure.  When I want to produce an HDR image, the exposures values for my images are usually -2EV, 0EV, and +2EV.  This means one normal exposure (0EV), one 2 f-stops darker than normal (-2EV), and one 2 f-stops brighter than normal (+2EV).

My camera has a "bracket" function where I preset the number of images I want, and the EV for each image in the bracket.  With the press of the shutter button, my camera takes the three individual images in rapid succession, each with the preset EV.  In post-processing, the software combines the three images into one image.  There are many software packages that you can use to make HDR images.  A quick Google search will identify them all, but a few notables are Photomatix, and HDR Efex Pro which is free from Google.  I personally merge the three images into a single HDR image, and then apply post-processing all with Adobe Lightroom.

The three images at -2EV, 0EV, +2EV

So why would you want to make an HDR image?  Unlike the human eye, cameras can't resolve as much "dynamic range".  The image in the middle is 0EV exposure (the image my camera would take if I did not set it to take the other two images).  The darker one to the left is -2EV, and the one to the right is +2 EV.  Take a look at the top part of the middle "normal" picture.  The clouds and the top of the castle are a little over exposed and some detail and contrast are lost.  Now look at the bottom of the picture.  It's a little under exposed and it's hard to make out the full reflection of the castle. 

To make an HDR image, you take three images and combine them all together into a single image.  Later through a software process, you can bring up the shadows in the dark areas and bring down the highlights in the over exposed areas.  This gives you a single HDR image where you can see detail in what was the overexposed and underexposed areas.

The single HDR image without any other post-processing

I said I would not get too technical, but here I go, just a little.  So far we've been talking about "dynamic range".  When you combine the three images as exposed above, you get a single image with three times the amount of information for the software to work with versus a single image.  What that means is that you have more "range" and control for all your post processing choices, not just exposure, but  contrast ratios, tone mapping, etc.

So the two main reasons I like HDR images are when I take pictures of a scene with both very bright and dark areas, and/or when I want to have more control over the other traditional post-processing controls allowing me more flexibility of the final HDR image.

Here is the final HDR image