HDR

You are currently browsing articles tagged HDR.

HDR using Photoshop vs. Photomatix

May 4, 2009 in Photo, Photography, Software, Technique | No comments

Processing multiple exposures into a single HDR (high dynamic range) image turns out to be more of a subjective process than I initially thought.

Have a look at the following image, which was processed using Photoshop CS4 and the “Merge to HDR” feature:

Image processed from 3 exposures by Photoshop CS4 "Merge to HDR".

Image processed from 3 exposures by Photoshop CS4 "Merge to HDR".

Not bad, really, at first glance. But notice at the right edge, about halfway down the image, how the arch coming in from the wall is completely blown out?

Now look at the same source exposures converted by Photomatix:

3 Exposures processed by Photomatix Pro.

Image processed from 3 Exposures by Photomatix Pro.

The highlights are not blown out, the tonality of the image is smoother, the reddish cast is reduced. All in all, the image is much closer to the intended look than what Photoshop allows.

Given that Photomatix allows me a lot more control to tweak the result than Photoshop, I think the money is well spent on Photomatix …

Church Ceiling

April 27, 2009 in Photo | No comments

Church ceiling

The ceiling of the Freising cathedral.

The previous cathedral burned down almost exactly 850 years ago (April 5, 1159). Work on the current church started almost immediately but it took almost 100 years until the church was finally consecrated.

The interior was remodeled a number of times with the current rococo ceiling having been created by the Asam brothers around 1724. Everything looks shiny and new because major renovation and restoration works were completed recently.

The difficulty in capturing a scene like this is that the light coming in the windows is extremely bright while the shadow areas can be very dark. Most digital cameras can not handle the dynamic range in such a scene.

The solution is to use HDR (high dynamic range), which is a technique to squeeze more extreme light and darkness into an image than the camera sensor can capture. The trick is to create several exposures of the same scene (a tripod really helps :): one frame exposes the highlights correctly, the next the midtones, the final frame the shadows. Software magic then combines everything into one image. Because most monitors are 8-bit-per-color-channel devices (and JPEG supports only 8 bits per color channel) the resulting image is then rendered down into an 8 bit per channel representation.

The difficulty lies in the number of choices that must be made at every step. It takes a bit of experimenting to get good results.