Dialing In Skin Tone

tonalControl-Opener

Getting skin tones right is critical. When you’re shooting, you can be sure your tones are good by using the D-SLR’s histogram in Luminance mode. You can see that although the histogram is stacked to the left due to the dark background and the dramatic lighting, the midtones, where skin is represented, are looking good. This is a tricky image to evaluate with the histogram.

For fashion, beauty, portrait and wedding photographers, the advent of digital photography has been a godsend. For wedding and portrait photographers, in particular, digital has enabled many professionals to increase productivity, thanks to faster turnaround times and greater flexibility. By reducing the time to delivery, digital photography has allowed the wedding photographer to capture the bride and groom’s desire for immediate fulfillment (pun intended) and sell more images while the event is fresh in the mind. Others have turned to on-site delivery, cranking out proofs and takeaway prints for guests, capturing a market that would have been impossible with traditional film photography.

But for all the power digital has to transform a business, the technology requires a greater degree of mastery, digitally speaking. Because of the way that digital sensors create their images, the resulting photographs can lack a large portion of the tonal range that their film forefathers would have easily recorded.

For the wedding photographer—where capturing every subtle nuance in a lightly colored wedding dress while still picking up the details of the piping on a tuxedo is crucial—understanding how best to craft an image for the digital medium is key. It’s important to understand not only how digital cameras differ from film in terms of image capture, but also how to best set up a shot to capture the greatest range of tones.

Bits Of Buckets

Every photograph, be it from film or a digital sensor, has a certain range of tones that can be produced, and the distribution of lights to darks is called dynamic range, which is commonly expressed in ƒ-stops. A typical piece of print film may yield a 10-stop dynamic range, while many digital cameras capture only around four stops of tone between the lightest and the darkest regions of the image, with anything outside those four stops resulting in the image being clipped to white or dark. (One of the selling points of medium-format digital backs is that they boast dynamic range up to 12 stops, which is why they’re so often used in high-end portrait work.)

High-school physics classes introduced the idea that light acts both as a wave and as a particle. It’s the particle properties of light that are important to the wedding photographer. Across the face of a digital sensor are light-gathering areas that work like little buckets, gathering up the incoming photons. If the bucket is empty, the region is black; half-full, and it’s a midtone; and if completely full, it’s white.

Here, again, we have an image that’s tough to get a handle on with the histogram. The whites of the veil are quite hot, but the skin tones look very good.

The problem for digital photographers lies in the size of the bucket—as more pixels are placed on the face of a sensor, the buckets have to get smaller and smaller to accommodate all the buckets. When a bucket fills up, the resulting part of the image is clipped to white (meaning it loses detail) even though there may be more photons available. A photo carries information, so a bucket that fills at 10 photons clips to white before a bucket that fills with 20.

This is an oversimplification of the process, of course, but it illustrates one of the common problems with digital photography—as more and more pixels get pushed onto digital sensors, the range of detail available increases while the dynamic range decreases. Since the range of tones is limited by the size of those photo buckets, improper exposure can dramatically limit the amount of contrast in an image.

Rules Of Thumb

Many photographers use a rule of thumb for shooting digital that’s slightly inaccurate, but it’s a good on-the-fly guide for quick digital photography: Shoot your images slightly underexposed. The reason for this logic lies again in those buckets of photons—if you overexpose your images, you’re going to clip your image and lose detail in the highlights (and the human eye is better at seeing details in highlights than in shadows). If you’re slightly underexposed, you’re much more likely to preserve the full range of tones in your image.

iStock-closeup-lg

However, this rule causes issues when shooting particularly dark scenes, where you don’t have bright highlights and will instead lose detail in the shadows. Ideally, you should meter each image as close to the edge of your highlight exposure without clipping anything.

If the brightest tone in your image is a midtone, then you’re best to meter and expose for that as your highlight, since that will distribute the greatest range of tones across the rest of the image.

Most photographers do this by looking at the in-camera histogram of their image and metering so that the right edge of the histogram tapers off at the edge of the clipping point. However, there’s a caveat to this. Many cameras display their histogram information from the JPEG settings, and JPEG images can clip data before the RAW file would lose the data. That’s why it’s a good idea to shoot weddings and many other subjects using a handheld light meter.

Meter Made

While the exposure meters in digital cameras are extremely powerful, they’re not perfect, and the wedding photographer especially needs to maintain perfect exposure. That’s why so many pro wedding shooters still rely on the same handheld light meters that were so in vogue in the days of film.

A handheld light meter provides a measurement of the incident light falling on a scene, while the in-camera meters are reflective. Measuring the light hitting a surface is particularly useful because it eliminates the possibility that the camera’s meter is fooled by specular highlights. Bright areas of a scene like the sun, lights and other elements won’t throw off the metering.

Here’s why that’s crucial: If your sensor has only four stops of dynamic range, and the camera’s meter sees a bit of light reflected off of a mirror as being the average highlight, but it’s actually two stops brighter than the white of the bride’s wedding dress, then two stops of dynamic range have been lost to that mistaken metering.

If you properly meter for the highlight of the dress (ignoring the reflection in the mirror), however, then you’ll end up with four stops of detail below that level, preserving all the scene’s contrast.

Histograms

By examining the in-camera histogram, it’s possible to get a better idea of the tonal distribution in the image. Your histogram (set to Luminance) should ideally be a bell curve with no spikes or gaps, and without having either end of the histogram cut off.

D3X_backbottom1-lg

Still, a scene that appears properly metered but suffers from a clipped highlight or shadow needs to be reshot, as it won’t be possible to recover data that has been clipped. It’s possible, however, to adjust contrast with software.

Postprocessing

While it would be easy to write an entire book on the subject of tonal correction in digital images, most programs try to simplify the process. They have a variety of tools (highlight and shadow, exposure, contrast, etc.) that all employ different ways to manipulate the same set of information.

Simply put, there’s a fixed amount of tone to go around in an image, but the relative levels of that tone can be adjusted as long as the sum still comes out the same. In other words, if two tones are really a stop apart, you can tweak things to make them half a stop apart, but that will result in changes to the overall tone.

Looking at a clean histogram, you can see that the tones are clumped toward the midtones, and there’s a good deal of space on the ends that weren’t utilized. This image would have more dynamic range available to it.

Using Aperture (although you could do this in Photoshop, Lightroom and other programs), you can make an adjustment to the exposure, which moves the tonal curve to the right of the histogram, which has the effect of changing the midtone of the image.

At this point, you’ll see that the histogram begins to clip on the high-lights, although it’s only clipping on the display; the information isn’t thrown out by adjusting the exposure.

Next, taking the Recovery slider, (which doesn’t really recover so much as bring back into range), drag it until the bell curve just ends at the right edge of the histogram. You can hit Option-Shift-H to see an overlay of the image displaying the hot (overexposed) and cold (underexposed) areas. This has the effect of compressing the tonal curve on the right side and redistributing the tones, at which point the midtones and the highlights have been brought into range.

Finally, grabbing the Black Point slider will remap the point in the tonal curve where shadow becomes solid black. Be gentle with this side of the curve because you don’t want to lose detail in something like a tuxedo or dark-colored bridesmaid’s gown.

These adjustments are analogous to using the Levels tool, but are done without having to fully understand the histogram. Experienced Photoshop users can bring up Curves and adjust the relative points of the tonal curve, while Aperture users can use tools like Shadows and High-lights to adjust the tonal range at individual points.

You can see more of Schloss’ photography by visiting his website at www.davidschlossphoto.com.

Leave a Reply

Menu