Inverted composite channel luminosity mask

Luminosity Masking 2

Controlling Edits with Intricate Masks

In my earlier tutorial on Luminosity Masking, I introduced the use of this type of mask and briefly described using panels and sets of actions from developers to make them more accessible to us. Although I showed some screenshots of luminosity masks, I didn’t emphasize criteria when it comes to deciding when to use them. In this article I’ll take a look at sources for luminosity masks within Photoshop (free), and point to the more obvious times when you would, or likely would not, use a luminosity mask.

I’ve already discussed the luminosity mask you can make by Cmd/Ctrl-clicking on a channel thumbnail (the composite and individual color channels). This loads only a lights/darks luminosity mask, with the lightest lights by default fully selected, progressing through a range of ever darker grays until finally the blacks are fully deselected. One of the more common uses for this mask is to invert the mask when tinting a B&W photo, thus tinting the darker values the most, while leaving the lightest tones light and bright. Often tints applied to an entire image muddy the lights and reduce the full tonal range of the image—often effective when going for the aged and distressed look, but easily avoided when you want to tint a “clean” B&W.

Tint results with and without a luminosity mask
For an aged and distressed look, you may want to tint all the tones in a B&W (top). To create a tint that preserves the tonal range, whether on restored photos or new B&W’s, an inverted luminosity mask (middle) does the trick. Cmd/Ctrl-click on the composite channel thumbnail in PS, then choose Select> Inverse to tint the shadows and protect the highlights by masking a Solid Color adjustment layer in Color blending mode.

Many people are not aware that Photoshop also includes a more precise method for generating a luminosity mask without using the Channels panel. When Photoshop updated the Color Range dialog, they added sliders to the Shadows, Midtones, and Highlights selections. The sliders now allow you to set a limit to the range of values considered Shadows, for example, and to decide how the selection will taper off. You’re targeting luminosity, not color, in your image, and generating the mask from that.

Select Color Range luminosity mask previews
One delimiter for Shadows, and one for highlights, adjusts how far the selection goes into the quarter or midtone range. The Midtones slider has delimiters for both an upper and lower limit to the range. The Fuzziness slider tapers off the selection so you won’t get harsh edges.

To know what tones to select, it helps to use the Info panel to read the values under your cursor before entering the Color Range dialog.  After opening the dialog, select one of the three basic values and use the sliders to select the values you’d noted earlier. You can also look at the little preview in the dialog to help you visualize what is being selected. Unfortunately, I do mean “little” preview. There’s no preview in the document itself, which makes it more difficult to be precise about your selection visually. However, a Color Range mask does give you quite a lot of control over limiting adjustments to a specific tonal range.Original Fall color photoLayers that use Color Range luminosity masks

Final Fall photo edited with Color Range luminosity masks
Applying Curves, Levels, and the USM filter to the top image through Color Range luminosity masks (middle), heightens the brilliance of the Fall colors (bottom) without overpowering the delicate detail.

Because choosing the right tonal range to target with a luminosity mask takes some practice, whether using Color Range, a panel such as Tony Kuyper’s, or directly using the Channels panel to create the mask manually, it helps to have a pretty good idea if your image is likely to benefit from a detailed mask. I’ve come up with three images that have a similar subject. They’re all a simple mountain range, but they vary in the distribution of values and their level of detail, as well as aesthetic considerations when editing the photos.

In the snowy mountain photo, there’s considerable detail and variation in values throughout most of the image. If I want to increase contrast through the midtones without blowing out the lights or blocking up the darks, a luminosity mask should be successful in limiting a simple Curves adjustment. No tedious hand painting is necessary.

Original photo and composite channel mask
The original image and the basic composite channel’s luminosity mask reveal detail and a mix of values throughout the image—a good candidate for editing through luminosity masks.
Final results after using a TK4 midtone luminosity mask
Choosing one of the Midtone masks in TK4 let me adjust most of the mountain range with a single mask without pushing any values too far. The mask protected the lights and darks more than the midtones from the contrast curve.

In the photo with the bird where the mountain range is nearly all one value, it isn’t particularly useful to try to find a luminosity mask that affects only the clouds and mist, and another to affect the contrast in the mountains themselves. There’s so little contrast and detail to begin with, it’s easier to add a boost with a standard  “S” curve, then broadly paint out the sky that has now been lightened more than desired. Another broad brush tones down the heavy darkening in the foreground mountain.

Original featureless mountain range
The original is really just a series of almost featureless shapes with very little contrast in values.
A simple handpainted mask controls the contrast curve.
After adding a simple contrast curve to the whole image, the mask was hand painted to prevent the sky from blowing out and the lower mountain range from overpowering the delicate wisps behind it.

The third image of heavy fog weaving between the mountains is bit trickier. There’s certainly enough detail and distribution of values throughout the image to make us think of using luminosity masks. However, I think the image will be more successful, and have more depth, if the front of the range draws the eye with greater contrast, leaving the rest to recede.  If I used a luminosity mask to target the midtones, all of the mountains would be affected. Since I only want to alter the lower third, it’s simpler again to apply a global contrast curve, then take a broad brush and paint out the effect on the rest of the image. You can do the same thing if you use a luminosity mask and paint on it to hide most of the effect. But why make an easy edit more complicated if you don’t need to precisely target a range of values?

Original photo of fog-filled mountain range
Although there is detail and contrast in values distributed throughout the image, which could make it a good candidate for using luminosity masks, enhancing all of that detail equally won’t help create depth or focus our attention anywhere.
Edited image with simple handpainted mask
Bringing out the foreground mountain, while allowing the rest to recede only requires a very simple mask.

Like many image editing features, luminosity masks are both easy to generate and very powerful when controlling what your adjustments affect. But also like many powerful features, it does take a bit of practice both in determining when to use them and what tonal range to use. Next time, I’ll demonstrate how using the TKActions panel helps when making those creative decisions. Layers edited using TK4-generated luminosity masks

Final photo edited with TK4 masks
Using the TK4 Actions panel rather than Color range generates different results—although difficult to see at small sizes. These masks are previewed in the full document and allow for more precise targeting of the entire tonal range.
Advertisements

2 thoughts on “Luminosity Masking 2”

  1. Let’s see if I have this straight. Luminosity masks work best on images with high contrast or those with a large dynamic range. Is that right?

    Like

  2. Luminosity masks work on almost any type of image, but they’re most useful when there’s a lot of detail scattered throughout a fairly large region. We see detail because of the contrast in values—light tones next to medium and dark tones, not grey next to barely lighter or darker grey.

    A pier jutting into a smooth lake can be minimalist and may have very little detail expressed with very few values. It can be high contrast, though, and have a large dynamic range.

    But if you see a close up of the pier well enough to see the wood grain, there are people on the pier, boats tied up to the pier, the waters are choppy and the sky filled with little clouds, you’ve got a lot of detail—a lot of values in small areas butting up against other values in small areas of the image. If you want to increase contrast, but not everywhere, you’re going to have to create a mask some way, and the easiest way I know of when you have all that detail is to generate a luminosity mask that targets the midtones, for example, but leaves the shadows and highlights alone, even though they’re intermingled with the midtones.

    Does that make more sense?

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s