Wednesday, January 28, 2009
Master high dynamic range imaging to enhance tonal range or create hyperrealistic effects
When the source images are combined into a single HDR image, the information from each image is analyzed based on both the low-bit pixel values and the EXIF data to create the HDR histogram. And in the 32-bit HDR space, all the black and white clipping from the source images disappears; we now have a single image with all of the information from the source images, with tonal values relating much more closely to their real-world values. There’s just one little problem: These HDR images contain more information (contrast ratio and dynamic range, for example) than our computer monitors and printers can display.
THE 32-BIT CHALLENGE
True HDR monitors are slowly creeping into the exceedingly high-priced specialty marketplace, but for the rest of us, our standard display devices simply can’t display all of the information in an HDR image. And despite advances in printer gamuts and some clever marketing terms, there’s simply no way to print a true HDR image. (Think about it like this: I can draw a picture of a lightbulb or fire, but my wax scratchings on paper can’t actually emit light.) Our monitors can display only a range of the pixels in the image around a given luminance value. Adobe Photoshop addresses this with the White Point Preview in the HDR Generation window (and the Exposure slider in the main Photoshop interface when 32-bit images are loaded). Slide the White Point Preview arrow, and the exposure shifts up and down, revealing pixels at different luminance values around the active white point. It’s important to note that no information is lost when adjusting white point—it’s just outside the visible slice of the image.
Save these merged 32-bit files. We’ll see HDR monitors in the not-too-distant future. But for now, we need to compress the 32-bit data into 8- or 16-bit space to display online and print.
THE ART AND SCIENCE OF TONE MAPPING
Crunching the HDR data into smaller bit depths is called tone mapping, and it’s kind of, but not exactly, like processing a camera RAW image.
The information in the 32-bit HDR image needs to be converted into 8- or 16-bit space for display in our traditional display methods. The tone-mapping process can produce amazing results to preserve luminosity, accentuate detail and crank up contrast on all scales. When the series of bracketed source images captures the entire tonal range of the scene, less software effort is needed to crank up details and contrast while preserving the luminosity of the scene—resulting in crisper, higher-quality images that can range from photorealistic to surrealistic without looking excessively ersatz and amateurish. And once you tone-map your image, you have a normal digital image file, but with the illusion of much greater tonal range and detail than would have been possible with a single shot. You might not need the HDR workflow for every shot, but there are times when going the HDR route is the best way to capture all the detail—whether it’s a cathedral in Rome, a famous bridge in New York City or wherever your photographic journeys take you.
Page 2 of 4