In-Camera Bracketing vs Exposure Correction in DPP

Question

Is there any difference in images produced using in-camera bracketing (-2,0,+2) and images produced using exposure correction (-2,0,+2) from RAW images in DPP? I'll be traveling tomorrow to a place where I think I'll take a lot of bracketed photos for HDR processing. Since I don't have too much memory cards with me, can I just take one properly exposed photo and later generate -2 and +2 exposed photos in DPP from it instead of taking 3 bracketed shots and get the same quality? Bracketed/Generated images will be used for HDR processings mostly, so its okay as long as the generated HDR (I use Photomatix 4.0) has the same quality and look.

[EDIT: I've done the test, the HDR generated from 3 in-camera bracketed shots gave me a tack sharp HDR with no noise but the HDR from exposure corrected images in DPP generated much more noise in the merged HDR, it was almost unusuable. I used Photomatix 4.0]

Answer 1

Exposure correction is using a single image to do something like HDR, but not true HDR. The reason is that you do not have the same dynamic range as three images provides. A true HDR will utilize multiple images, and have much more detail than a single image...meaning you will be able to see many more objects from the shadows of a single image, and you will be able to recover objects that are in the blown out detail of a single image.

When you do bracketing, the camera takes 3 images. One is exposing the entire scene two stops less than the 'base' exposure, but this will be correct for many objects in the image, and incorrect for others. Next, it takes exposures at 0 (base) and +2 (2 stops over), where you are exposing different portions of the image "correctly", and different portions "incorrectly". (I am using "correctly"/"incorrectly" in this instance to mean that there is adequate detail, vs incorrectly, meaning there is no detail available). In this case, each image has detail from different portions of the image that will be missing in the other images. When you combine the images, you are generating an image that combines all the exposed areas, throwing out the incorrectly exposed ones, thus leveraging ALL the detail available in all 3 images.

When you create one image, you expose for either an average of the scene, or a small portion of the scene, depending on camera settings. In this case, there are areas that are exposed incorrectly, and provide NO detail. In a single image, there is NO detail to be recovered, even in RAW...you simply will not have the information in your image.

The end result will likely be very different, however sometimes a single image "HDR" can provide more detail than you would get otherwise, and may be the photo you wanted. Just remember, you can not replace detail that simply isn't in your image, so its best to capture the detail as best you can, which means 3, or more images to capture all available detail. Also note that true HDR works best with RAW images.

Answer 2

The short answer is: It depends entirely on your ability to expose one photo correctly for the environments you'll be shooting. If you can, go with it HDR, if the environments truly do have more dynamic range than your camera can handle, go with bracketing.

The longer answer is not so clear - bracketing will generally give you more detail than just one exposure since you'll likely want to use either method in cases where you will exceed the dynamic range of your camera and HDR really is just a method of trading local contrast in to emulate a higher dynamic range. As the +2 and -2 images in bracketing will contain far more information in the highlights and shadows than just a RAW image exposed for midtones, you will potentially get more data to work with, all depending on the jpeg performance of your camera and your settings.

To get the most starting data for HDR, you'd probably want to go with bracketing RAW images, but considering that you want to preserve your space, that is not what you would want. If you use bracketing correctly, making sure to keep your depth of field constant, you will likely achieve far better images with bracketing.

Answer 3

You can just test it for yourself:

1. Find a location near your house where you can see both bright sky and deep shadows.

2. Take the 3 bracketed pictures.

3. Make two HDR images, one with the bracketed pictures and one with the images generated from just the middle image raw file

This will tell you if the single exposure HDR is good enough with your specific camera, lens, software and your taste (I suspect the multiple exposure HDR will be better - but the single exposure image can be good enough).