Fixing dark prints with a Darks-1 luminosity mask

I recently acquired a new Epson SureColor P800 printer. My old printer had started to leak from the print head. I primarily use Moab Lasal paper with a luster finish, so I went to the Moab website to find a color profile for the new printer/paper combination, and was pleased to find that one existed. I loaded it on my computer and did a test print. The color match to my monitor looked great. There was a problem, however. The print seemed dark and heavy overall because the shadows were blocked when viewed in ambient room light. Looking at the print under a brighter light source, I could still see the shadow separation that was apparent on my monitor, so it hadn’t been lost entirely. It just wasn’t printing well. I don’t normally view my prints under strong “gallery” lighting and want them to look good even in ordinary room light. So based on this initial test, I would need to find a consistent way to lighten them. The fact that there was NO color shift in the print meant that the profile I installed was doing a decent job in that important category. Brightness was the only thing that required adjustment.

I experimented with a few more prints that had Curves and Levels adjustments added to the sharpened image, but these ended up with obvious shifts in contrast and color saturation. I then decided to see if a luminosity mask could help. Since the image’s dark values were blocked, a Darks-series mask would be a good place to start. But how to use it? Curves and Levels adjustments hadn’t worked, so I needed an alternative for that also. In the end, I decided to still use an adjustment layer, but instead of adjusting the properties for the layer, I simply changed the layer’s blend mode to Screen, which lightens the image, except for pixels that are completely black.

Bingo! This approach worked well and next print looked much better. Here’s the steps I used:

  1. Create a Darks-1 luminosity mask of the unsharpened image. (See NOTE at the end of this blog for methods to make a Darks-1 mask.)
  2. Sharpen the image using the normal sharpening method.
  3. Create a Levels (or Curves) adjustment layer at the top of the layer stack above the sharpening layer(s).
  4. Set the adjustment layer’s blend mode to Screen.
  5. Apply the Darks-1 luminosity mask created in Step 1 as a layer mask to the adjustment layer.
  6. Lower the layer’s opacity to between 25 and 50 percent.

The image below shows the final layer stack with the print adjustment layer on top.

workspace menu button

Not hard at all, and my image now matched my monitor in color, contrast, saturation, and brightness. In retrospect, this is sort of an obvious solution. The darks were blocked and Darks-series masks specifically targets these dark values. Screen blend mode is also a somewhat obvious choice since it provides an automatic, consistent lightening adjustment that keeps 100% black pixels 100% black, so there is a bit of a black anchor for the blackest blacks in the image. Screen blend mode also seems to lighten with less contrast and/or saturation shift that sometimes accompany adjustments to Curves and Levels using the Properties panel. So the overall effect was to create an automatic lightening effect restricted to just the dark, blocked tones in the image.

Subsequent tests showed that the optimal opacity of the added adjustment layer ranged from zero to 50 percent. It turns out that the value is easy to predict based on image’s histogram.

For images with prominent areas that are very dark or completely black, an opacity setting of 50% is needed.

workspace menu button

For images where there is plenty of dark colors, but still excellent detail in these dark colors, an opacity of 25% works well. This histogram shape is common with many of my images, so an opacity setting of 25% is frequently my starting point when making test prints.

workspace menu button

For images that are composed almost entirely of midtones and/or highlights with very few dark values, no adjustment is needed. There are essentially no dark values that are getting blocked in this situation, so the adjustment isn’t necessary.

workspace menu button

While a custom profile might have been an alternative to consider, the fact that the colors were matching so well with the paper manufacturer’s profile meant that I wanted to continue to use it if I could. So far I’ve made prints of about a dozen different images using this method, and opacity settings of either 50% or 25%, depending on the histogram, have worked every time.

I’m not sure how widespread this problem might be. A member of the local camera club mentioned he had the same issue, so I’m sure I’m not the only one. It’s certainly an easy fix to try if your prints are looking too dark.

NOTE: A Darks-1 mask can be made and applied with either the TKActions V6 RapidMask2 module or the free Basic V6 panel. You can also make a Darks-1 layer mask using Photoshop’s Image > Apply Image… menu command using the settings shown below. Be sure to check the “Invert” checkbox in the dialog window to get a Darks-1 mask instead of a Lights-1 mask.

workspace menu button

TK Actions Quick Tip: Three ways to use Levels and Curves

There are a lot of features packed into the TKActions V6 panel. Not only does it quickly generate luminosity masks, but it also allows them to be modified and output in a variety of ways. In addition, the V6 panel also provides access to most of the common Photoshop functions photographers use to develop their images. Curves and Levels are a natural part of many techniques, so different V6 modules contain buttons and menu items to appropriately access these adjustments. In this new Quick Tip video, Sean Bagshaw reviews the locations of the Curves and Levels options in the V6 panel and demonstrates how each can be used.

There are three distinct Curves and Levels adjustments found in the V6 panel:

  1. In the “Layer Mask” menu of the Combo/Cx modules. Buttons in this menu are designated by the familiar Photoshop icons for these layers. They create the corresponding adjustment layer with a white “reveal all” layer mask if there is no active selection. If there is an active selection, the selection is incorporated into the the layer mask as the adjustment layer is created. These buttons work similarly to the items in the “Layer Mask” menu accessed at the bottom of the Layers panel. However, there is one advantage to using the Combo/Cx buttons and that is that they automatically open the Properties panel after the adjustment layer is generated. Most adjustment layers do nothing until you adjust the layer’s properties. Anticipating this, the panel opens the Properties panel so users can go directly to making their adjustment after creating the layer.
  2. In the MODIFY section of the RapidMask2 module. The RapidMask2 module is all about making pixel-based masks, like luminosity masks. While the panel can make dozens of different calculated masks as a starting point (Lights, Darks, Midtones, and Zones), there are actually an infinite number of possible masks for any pixel-based value. The MODIFY section allows you to fine-tune any mask to better match the pixels in the image, and Curves and Levels adjustments fit nicely as one of the modification options. The MODIFY buttons open a separate window where users can watch the mask change in real time as the adjustment is made. There is also a MODIFY section in Layer Mask Mode, and Sean’s video demonstrates how Curves and Levels work in this section as well.
  3. In the “Layer” button menu in the OUTPUT section of the RapidMask2 module. This is the final location for Curves and Levels in the V6 panel. “Layer” menu items automatically apply the current Rapid Mask that has been created by the module to the adjustment layer that’s created. It’s a one-step process to go from mask creation to being able to actually use the mask to adjust the image. Besides the efficiency of using this output process to deploy masks, there are two additional advantages. The first is that no intermediate 8-bit selections are used. It’s a direct channel mask to layer mask process, which is 16-bit to 16-bit. The quality of the layer mask is therefore identical to that of the original Rapid Mask. The second advantage of this output method is that the Properties panel for the new adjustment layer once again automatically pops open. Adjustment layers need an adjustment by the user, and this output method lets you get right to it.

Sean, as usual, does a great job reviewing and demonstrating these different options. I’m sure you’ll feel more confident using them after watching this video.

Be sure to subscribe to Sean’s YouTube channel for more great tips on photography and post-processing including those listed below.
Quick Tip: Three ways to use Levels and Curves
Quick Tip: Reusing saved luminosity masks
Quick Tip: Developing a quality night sky
Quick Tip: Split toning
Quick Tip: Cloud sculpting
Quick Tip: Exposure blending
Quick Tip: Favorite new V6 features

TK Actions Quick Tip: Reusing saved luminosity masks

Sean Bagshaw recently posted another Quick Tip video covering how to save and reuse luminosity masks. The TKActions V6 panel has the “Channel” button in the OUTPUT section, which saves the luminosity masks it generates, already programmed in. So the saving part is pretty easy. But the panel doesn’t know when you want to actually recover and use the saved mask, and that’s where Sean’s video comes in handy. He’ll walk you through a couple of different methods to add the saved mask to a layer on the Layers panel. Saving and reusing masks on the Channels panel is somewhat of an advanced technique, but it’s helpful to be aware of it for those situations where you’ve spent a little time creating a particularly useful mask that you might want to use again later on.

Another option for reusing and duplicating a previous luminosity mask is to use one already deployed as a layer mask on the Layers panel. To reuse a layer mask, simply ALT/option+click on the mask and drag it to the layer where you want to use it. Photoshop will duplicate the mask and add it as the layer mask on the second layer. While this is a relatively easy maneuver, layer masks can sometimes get a little messy if you’ve painted on them to conceal or reveal how the mask affects the layer it’s attached to. Channel masks, like Sean demonstrates in the video, can be used to save an original version of a mask that can then be further altered with painting or other adjustments after it’s turned into a layer mask.

Be sure to subscribe to Sean’s YouTube channel for more great tips on photography and post-processing including those listed below.
Quick Tip: Reusing saved luminosity masks
Quick Tip: Developing a quality night sky
Quick Tip: Split toning
Quick Tip: Cloud sculpting
Quick Tip: Exposure blending
Quick Tip: Favorite new V6 features

TK Actions Quick Tip: Developing a quality night sky

Sean Bagshaw’s YouTube channel has another great workflow video using the TKActions V6 panel. This one covers developing the Milky Way in a night sky image. He starts in Light Room with what appears to be a somewhat unremarkable image of the Milky Way, but with a few quick adjustments uncovers the potential hiding in the dark tones. I liked the Light Room techniques for adjusting the colors in the RAW file and the way the blue fringe around the stars can be removed.

The really good stuff happens when the image is opened in Photoshop as a smart object. Sean selects the Blue channel mask to better target the Milky Way (compared with a standard luminosity mask). He then uses classic luminosity painting to create more dramatic contrast. This involves loading Light and Dark masks as selections and painting white and black through these selections to selectively change image brightness. It’s quick and easy since the selection guides the paint to where it’s needed.

For noise reduction, Sean uses the TKActions V6 panel to duplicate the smart object so the noise reduction can be done on a separate layer and later filtered into the image through a luminosity mask. Using the V6 to duplicate smart objects has the advantage of unlinking the duplicate smart object from the original smart object. This is important since Sean reopens the duplicate smart object in Adobe Camera Raw and is able to perform significant noise reduction on it without affecting the original smart object. If the original and duplicate smart objects had not been unlinked, both the original and the duplicate would have been affected the same by the noise reduction in ACR. Instead, by creating an unlinked smart object first, strong noise reduction can be applied to the duplicate smart object to the point of blurring the stars. A luminosity layer mask can then be applied that reveals the noise reduction in the dark areas of this layer while simultaneously preserving the sharp detail of the individual stars from the original conversion. It’s a bit hard to describe in words, but watching Sean do this in the video will make it crystal clear.

The final technique is to make the stars even sharper using the “Clarity” action in the V6 panel, again combined with a Blue channel luminosity mask as a layer mask. This is a subtle change but the type of one that makes the final image look its best.

The video is just over 21 minutes long, but it seems much shorter. Sean covers a lot of territory with a several really useful techniques facilitated by the V6 panel. The change is quite dramatic and shows what’s possible even when the original RAW file looks marginal. If you’re already doing Milky Way photography or thinking of trying it, this video will give you confidence that you can indeed create a stellar image.

Be sure to subscribe to Sean’s YouTube channel for great tips on photography and post-processing including those listed below.
Quick Tip: Developing a quality night sky
Quick Tip: Split toning
Quick Tip: Cloud sculpting
Quick Tip: Exposure blending
Quick Tip: Favorite new V6 features

The V6 RapidMask2 Module: Any mask, any time

I recently completed an updated video (below) on using the V6 RapidMask2 module. This module is at the heart of the TKActions V6 panel and was designed to be a mask-making juggernaut. Its evolution can be traced to the original luminosity mask concepts, but it’s moved far beyond the confines of those earlier techniques. One compact module can now make any pixel-based mask with just a few mouse clicks. Color, channel, saturation, and vibrance masks are as easily generated with RapidMask2 as standard luminosity masks. The built-in Rapid Mask engine quickly turns pixel values into masks, and these masks are viewed on-screen at near real-time speeds so it’s easy to experiment with different masks and find the best one.

It’s worth noting that all RapidMask2 masks are created using calculations, which provide the smoothest masks of any method to generate them. I experimented with a Curves adjustment layer for generating masks and even created a prototype panel using this method. However, I abandoned Curves when I saw the obvious tonal separation for tones with low pixel density in the image histogram. Calculated masks in RapidMask2 automatically adjust to match pixel density in selected tones by varying mask brightness. This isn’t possible when a static Curves adjustment creates the mask. So I’ve stuck with calculations for making masks in RapidMask2 and am confident it produces the best possible masks.

These calculations also completely avoid 8-bit selections as masks are generated and deployed. While I previously described the calculations process for making 16-bit masks and built it into the Rapid Mask engine, the reality is that calculations always make masks that match the bit depth of the image. Even 32-bit masks are possible with Rapidmask2 if you’re using the 32-bit mode in Photoshop.

The video below walks you through the workflow for using RapidMask2 to create and use pixel-based masks. It’s basically a four-step process:

  1. Choose a data source (luminance/color/saturation/etc) in the SOURCE section.
  2. Click different masks in the MASK section to find the best one.
  3. Optionally adjust the mask using the MODIFY section.
  4. Deploy the mask using the OUTPUT section.

This video will show you that it’s actually quite easy to make and use pixel-based masks once you have a panel that does most of the work.

More information on using RapidMask2 and the other V6 modules can be found in Sean Bagshaw’s V6 Video Guide series on the Panels & Videos page.

Monochrome 2: Toning with TKActions V6

The first article in this monochrome series dealt with using the TKActions V6 panel to convert color images to black-and-white. This second article looks at another important aspect of monochrome images: toning. Proper toning can enhance the mood of the image. Warm-tones, like sepia, impart a vintage look, while cooler tones communicate a more modern feel. There are a variety of different ways Photoshop can be used for toning. The Solid Color and Hue/Saturation adjustment layers when combined with luminosity masks are two of my favorite methods. The video at the bottom demonstrates both techniques.

Toning Theory

Historically, toning in the darkroom was a secondary development process. After the initial development to reveal the latent image on the exposed paper, the print was immersed in another solution that would chemically react with the silver in the print to produce the toning color. The areas of the image that had more silver (the dark areas) would react more strongly with the toner. Areas with less silver (the light areas) reacted less. This is an important distinction. The entire print did NOT receive equal amounts of toning. Dark areas received more and light areas received less. Visually, this meant that light areas of the image would still retain much of their original whiteness and didn’t change color dramatically in the toner. Very dark grays and pure black toned the most, but a dark shade of any color is hard to distinguish from black, so the color change wasn’t necessarily all that visible in very dark gray and black tones of the image either. However, as the grays in the image got a little lighter and on into the midtones, the toning color became much more obvious in the print. Grays in this range showed a definite color change, but still in proportion to the amount of silver in the print. A Zone 4 gray, for example, would show more intense color than a Zone 6 gray, and this was important to maintaining the tonal separation that created the image in the first place.

untoned image

toned image

This concept of applying toning to the silver-rich areas of a monochrome image (the dark areas) is where luminosity masks really come in handy. The Darks-series masks do exactly this. They select dark tones in the image in proportion to how dark the tone is. The darker the tone, the more it is selected. Using Darks-series masks as part of the digital toning process simulates what happens chemically in the darkroom. However, I’ve also found Midtones-sereies masks quite useful for digital toning. Since the midtones are often where toning is most obvious in chemically-toned prints, using them in digital toning also makes sense, and the results often look good. Which mask works best depends on the image. It’s helpful to try several different luminosity masks to find the best one. One more corollary to all this is that the best monochrome images for toning generally have some good gray midtones. Images with strong contrast where most tones approach black and white won’t show the toning color as well as images that have a lot of gray midtone values.

Toning vs. Tinting

As described above, toning was a chemical process that was strongest in the silver-rich parts of monochrome images. The effect is generally most obvious in the dark and midtone grays. The whites are largely unaffected due to their lack of precipitated silver. Tinting, on the other hand, can be thought of as a staining process that affects the entire image, not specific tones. With tinting, it’s the print substrate (the paper or emulsion) that is experiencing the color change. The midtone grays will still exhibit the color change, but so will the image’s whites. Both toning and tinting are ways to add color to the monochrome image, and in the digital darkroom, it’s possible to specifically target light tones as well as the darks and midtones. So in the digital realm, the distinction between toning and tinting is a bit less clear as it was in the traditional darkroom. One distinguishing characteristic is whether the color is applied uniformly across the image (tinting) or whether it’s restricted to specific tones (toning). The Photo Filter adjustment layer, for example, tends to be more of a tinting technique since it’s applied evenly across most tones in the image. For toning, Solid Color and Hue/Saturation adjustments tend to work best. The left image below shows toning accomplished with a Hue/Sat adjustment layer and a luminosity mask. The image on the right is a tinted image made with a Photo Filter adjustment layer.

tone vs. tint

Additionally, toning should be proportional within the selected tones where it is applied. The more selected tones should exhibit the most color change which then tapers off in tones that are less selected. Again, this is where luminosity masks really shine as they are able to isolate the color to specific tones in the image and in proportion to how much each tone is selected.

“Color” Blend Mode

Using the “Color” blending mode for the adjustment layer that adds the toning color is an additional technique that makes digital toning work well. It’s actually essential when using a Solid Color adjustment layer for toning, where it makes for a very nice toning effect when combined with a luminosity mask. It also provides a bit of a contrast boost when a Hue/Saturation adjustment layer is used for toning, so it’s definitely worth trying.

Summary: Toning is a way to add color and mood to black and white images. Solid Color and Hue/Saturation adjustment layers in combination with luminosity masks are a great way to digitally tone photographs that mimic the traditional darkroom process.

The video below demonstrates two good methods for using luminosity masks and adjustment layers to make digital toning easy and natural-looking with plenty of options for customizing the effect. If you want to practice along with the video, jpeg versions of images can be downloaded here.

NOTE: With the release of CC 2019 in October 2018, there is a blending mode bug on Windows computers. It can cause the wrong toning colors to appear and can also cause PS to crash. The workaround to fix this is to go to Preferences > Performance and make sure there is a check mark in the checkbox for “Legacy Compositing.” This post has more information about this bug and its current status.

Monochrome 1: Black and white conversions with TKActions V6

Monochrome is at the heart of the photography. Early photographs had one color, black (from precipitated silver), and, in combination with a lighter printing substrate (a copper plate initially, but eventually paper), created the classic black and white look that was the signature of photography. Monochrome images have evolved significantly since their origin in the 1800s, but they’re still widely appreciated and are a great way to interpret and view the light captured by a camera.

Masks in Photoshop are monochrome (black, white, and gray) by default. Luminosity masks opened the door to pixel-based masks where the monochrome tones in the mask have a direct correlation to the underlying luminance values in the image. Luminosity masks are not blobs of paint applied with a paintbrush run by a mouse. Their pixel-based nature means that the image itself can always be seen in the mask, at least to some degree. So there’s a definite connection between a luminosity mask and the image. And once you start working with luminosity masks, it’s hard not to notice that some of them could make good black and white images on their own.

As I coded the TKActions V5 and V6 panels, I made sure to include the option of utilizing masks as actual images. You never know when a surprisingly good mask (or at least interesting one) might show up, and only then realize that a monochrome interpretation of the image might be worth exploring. The V6 panel makes this easy to do with dedicated output options that preserve promising masks. The panel essentially functions as a color to black and white converter in this situation. While this isn’t the primary use of either luminosity masks or the panel, it does provide a convenient method to start experimenting with monochrome since the RapidMask2 module can quickly generate so many different masks. This article will focus on ways to use the luminosity masks generated by the V6 RapidMask2 module as the starting point for great black and white images.

Method #1: Mask To Pixels

The “Layer” button of the RapidMask2 module has a menu item for creating a pixel layer from the Rapid Mask. It’s called “Mask To Pixels” and is a direct 16-bit conversion from Rapid Mask to pixel layer in RGB Color mode. No image quality is lost in the process. The 16-bit pixel layer is an exact duplicate of the 16-bit Rapid Mask since there is no intervening 8-bit selection involved.

Mask To Pixels menu option

If the intention is to purposely convert the image to monochrome using the RapidMask2 module, it’s usually best to start with one of the “Lights” masks. Then you can optionally use the MODIFY section to further adjust the mask in order to optimize it before triggering the pixel layer output. In the cactus blossom image below, a Lights-1 mask was modified with a Levels adjustment to produce the final conversion.

lights mask to black and white

Alternate Method #1: Start with zone masks
One rather interesting way to use the “Mask To Pixels” option is with Zone masks. Zone masks are narrow slices of the tonal spectrum, but they’re still monochrome, and they are calculated to have smooth blending into adjacent tones. So there is a smooth tonal gradation as they transition from lighter grays of selected pixels to dark grays and eventually to black for unselected pixels. For simple compositions, like the cactus blossom shown here, try different zone masks or combinations of zone masks. Then use the “Auto” button in the MODIFY section of the RapidMask2 module to normalize the mask so that it has a full tonal spectrum from black to white. This can lead to some slightly more abstract but appealing black and white images. The strong contrast of the “Auto” command creates a wonderful silvery quality in the narrow range of tones selected by Zone masks. This shiny effect is difficult to achieve when working with images with a wide range of tones, but when working with a narrow slice of tones, it happens quite easily. For the conversion below, I added a Zone-7 and Zone-7½ mask together, modified the result with a Levels adjustment, and then used the “Auto” option on the result to obtain the silvery shimmer in the flower.

zone masks to black and white

Method #2: SOURCE > Color > Create

The “Create” option in the “Color” menu was intended as an easy way to make individual colors either lighter or dark in the mask that was being created.

color-create menu

The set-up for doing this requires a few different layers on the Layers panel. This set of layers also easily lends itself for converting images to black and white. So this “Create” option can serve a dual purpose. It can be used to create a Rapid Mask, which is output using the “Rapid Mask” button on the panel, or it can be used to convert an image to monochrome, in which case the correct output choice is the “B&W” button.

black and white button

The “B&W” option keeps the layers used to create the mask intact and simply renames the group that contains them. This provides a completely non-destructive way to do the conversion. At any future time, it’s possible to tweak the conversion by returning to these layers and making additional adjustments.

black and white conversion layers

Below is a color image and its black and white conversion using the Color > Create option.

color image
black and white image

Summary: Luminosity masks, because they are based on pixel-level data, embed the actual image in some manner in the mask. And because masks are always monochrome, luminosity masks have the potential to be a conversion tool for creating black and white images from color files. The TKActions V6 panel includes dedicated output options (“Mask To Pixels” and Color > Create) for using luminosity masks as actual black and white images. The video below demonstrates these techniques.

NOTE: Converting a color image to monochrome is usually not the same as actually finishing it. After converting to monochrome, most images will still require additional processing to achieve their best potential as a black and white photograph.