Dealing with Dust Using History in Photoshop

There’s another trick for dealing with dust in Photoshop (to sort of continue from the last blog post), which takes a little setup but is even more subtle. It involves the History Brush tool, and the History panel. The advantage is that the corrections are very unobtrusive, especially if one takes the time to do them carefully. The disadvantage, such as it is, is that the recipe has to be followed rather carefully, which is why I’m taking the liberty of condensing it at the end of the post.

After opening the file, it’s advisable to save it as a PSD, and make a copy of the layer containing the artwork; work on the copy for safety.

Gaussian Blur

The first real step is to go to Filter—Blur—Gaussian Blur, and set it for 6-12 pixels (all numbers approximate; your mileage may vary).

Then, in the History panel, take a snapshot of this state (call it Blurred, or whatever you’ll recognize). After which, click one step up in the History (which will un-blur the art but leave the snapshot blurred). THIS IS A KEY STEP.

History Panel

Click the selector for the blurred snapshot (left side of panel) to indicate what the History Brush should work from. Then select the History Brush itself in the Tools panel.

Blending Mode

THE OTHER KEY STEP is this: When painting a dark spot with the History Brush, switch the blending mode in the Control panel up top to Lighten, and vice versa. So you’re lightening and blurring the dark spots, and darkening and blurring the light spots. Lightening or darkening makes them less conspicuous; blurring blends the repair in.





Don’t panic if you need to practice this one a bit. The recipe is a bit more complex, and has to be followed. But I can tell you it works, as I’ve used it myself many times.


Short version:

Open file, save as PSD if necessary, make copy of layer

Blur image (Filter/Blur/Gaussian, 6-12 pixels)

In History panel, take snapshot and name accordingly

Go back one step

Select blurred snapshot

Use History Brush tool to paint blur back in (Lighten mode on dark spots, and vice versa)

How to Deal with Dust in Photoshop Images

In the age of digital photography, dust would seem to be a thing of the past. Perhaps. But who knows how many pre-digital pictures still exist, un-digitized, and in need of cleaning? With this in mind, I’d like to show you how.

Before we had digital cameras, Photoshop, and so on, one would have to clean the film very carefully, print the picture, and use a kind of watercolor called spotting dye to touch up the (hopefully few) white spots caused by dust on the negative. (Yes, I did it.) It’s not too hard, but it is tedious, and requires patience and a steady hand. Photoshop, of course, removes the need to do the spotting by hand, but how to actually deal with the dust spots?

Dusty Wedding Pic

With an image open, you could just borrow nearby color with the Eyedropper tool, then use the Brush to tap color into the white bits. This does work. But there’s a slightly more subtle technique which works better, is a little faster, and doesn’t need any eyedropper-ing.

Selecting Dust Spots

First, select a few spots (to start with, do one or two at a time). You want to make your selections only about half again as large as the white spots. Any selection tool is fine; with roundish spots, the Ellipse Marquee tool works well.


Then, go to the Select menu, Modify, and Feather command. We want to feather the selection by at least a few pixels, though we normally don’t need much more unless the spot is big. About three or four pixels’ worth will do.


Next, we go to Filter–>Blur–>Gaussian Blur. What we want is to “smear” the “paint” immediately around the spot onto it, but soften the edges of the smear to make it very gradual, and therefore hard to see. This is where there might be a little experimentation; the nice thing about the Gaussian Blur is, one can control how much blur there is. Sometimes it works with just a few pixels’ worth; sometimes it can take as much as eight or nine. The tricks are to keep the initial selection just a little larger than the dust spot, and feather the edge (either before selecting, in the Control panel, or right after with Select–>Modify–>Feather) prior to blurring. We do NOT want sharp edges here at all.

And don’t worry if it takes some practice to get right. Maybe make a copy of the layer containing the picture, for insurance. Then go for it.

Using the Layers Panel in Photoshop–Tips and Tricks

The Layers panel is one of the most important members of the team; a lot of what goes on in Photoshop is related at least partly to which layer(s) you’ve selected, or want to select to do stuff in. So paying attention to this, and knowing a few things, will save you some work, and a little stress.

Unlike in its brother programs, Illustrator and InDesign, layers serve a vital function here—anything in a Photoshop layer is basically oil paint that never dries. So keeping things on separate layers till you’re SURE you don’t need to is good procedure.

Before painting, clone stamping, erasing, or anything else, especially if you’re in the middle of an important project, glance over to see which layer you’re about to affect. You can easily select the right layer by clicking its name once.


If you want to merge two or more layers, you can easily select them, even if they’re not adjacent, by clicking the one’s name and control/command-clicking the other(s). Remember, merging layers normally sort of “collapse” such that whatever’s on top, in the middle, etc. will stay looking that way when done.

Layers Visibility

Turning off a layer’s visibility is an easy way to avoid editing it when you don’t want to. Click the eyeball on the left of the layer name. But don’t forget to turn visibility back on when you *do* want to edit.

Locking Layers

Locking a layer is a better way to avoid editing, especially if you want to see how the layer interacts with its buddies while working in the others. You can lock the image pixels, the transparent pixels, the position of the layer, or all of the above (usually the most common) with the buttons at the top of the panel. Just remember, you need to unlock before editing. Duh. 😉


Linking layers together provides a useful capability—if two or more layers need to be positioned just so, relative to each other, do the positioning, select them, and link the layers with the button at the bottom of the panel; if you then move one, you’ll move all. Unlinking them is easy too. Same button.


Renaming them is easy—double-click the current name, type what you like, and hit Enter.

Finally, remember Photoshop creates new layers whenever you copy and paste, or cut and paste, or pretty much anything and paste. It’s conservative about this, so you don’t smear the “paint” from different elements together. Which means you can accumulate a LOT of layers. In fact, as far as I know, there’s no limit to the number you can have. (I got bored and stopped counting at two hundred and twenty.) So don’t be afraid to merge them when you’re done with separate elements—they add to the file size. And don’t create them just for the heck of it—again, the file gets bigger when you do this.

Using Paths to Make a Selection in Photoshop

An unexpected ally in making selections in Photoshop is the Paths panel.

Selecting a precise piece of an image can be tedious, even for a veteran user; understanding the selection tools does not provide instant expertise, and some images have such irregular content that it’s a touchy matter to set the right numbers for almost any of the tools. But the Paths panel lets the user at least partly bypass the problem.

Using the Pen tool, we can create a path (essentially a vector shape) within Photoshop. This also takes practice, but the nice thing about a path drawn with the Pen tool is that it can be modified almost indefinitely. (By analogy, I’ll date myself badly and mention Silly Putty.)

Basic Paths

First, we bring up our image, select the Pen tool, and start creating anchor points etc. for the path. This can take some time—in fact, it’s a good idea to go fairly slowly—but since we can clean up later, a rough shape will do. Naming the path is usually a good idea for reference, especially if you’ll have more than a couple.

Cleaned Up

The next step is to use the other vector tools to refine the path as needed: The Add and Delete Anchor Point tools to give the shape more or fewer anchor points, the Direct Selection tool to fine-tune the locations of the points and tweak the control handles, and occasionally the Convert Anchor Point tool for changing curve to corner, or vice versa.


Finally, we can select the path in the panel, go to the panel menu, and click Make Selection. We then have to specify the various options in the dialog box, and click OK.


This is one of those not-technically-complex-but-kinda-tedious features which requires a little practice. Working with Photoshop’s brother program Illustrator would give some additional footing on this too. Couldn’t say whether it’s the *best* technique, but it seems to work well with practice, and it’s not too hard to fine-tune paths.

How to Use Layer Styles in Photoshop

The use of Layer Styles (also known as Layer Effects) in Photoshop is fairly easy, but can add some pretty snazzy results. One of the most visually catchy is the Drop Shadow, which can make it look as if an object is in front of the rest of the artwork. Any of the styles can be applied in about the same way, but the Drop Shadow gives a good general example.

Layers set up

Usually the first thing is to assemble the artwork layers, since the effect is normally applied to the whole layer. As far as possible, get things positioned (though moving layers and items around does not mess up the layer effect).

Layer Effects box

Double-click  the layer thumbnail to bring up the Styles/Effects box, or select the layer and go to the Layer menu/Layer Style submenu and select the effect you want to apply.

At this point, the controls will vary widely according to which style you select, but here’s one point you don’t always hear about: If you want to open the box by double-clicking the layer thumbnail, remember to then click not the check mark next to the effect name, but the name itself. Why? Because if you click the check mark only, it will activate the effect but NOT bring the controls up. You’re saying you want to accept whatever defaults are already there. And usually, the user is looking for a particular fine-tune of whichever effect.

For our example, I want to make it look as if the gull is flying in front of the picture, almost as if he had somehow gotten free of the poster. So I can set the angle (from which the light comes) as 125 degrees (upper left), the distance (of shadow from object) at 150 pixels (so I can see it clearly as a shadow), the spread big enough to keep the edges soft, and the size (which is actually the blur) fairly large, since most shadows in this situation are not sharp-edged.

Drop Shadow effect in

Click OK, and it’s done. The Layers panel shows the Fx symbol, to let us know there’s effects in them there hills, and we can double-click the thumbnail or the Fx anytime to get back in and play around with it some more, or delete it completely and start over. Fully editable. Way cool.

Creating Alpha Channels in Photoshop

Making selections in Photoshop is fairly easy, but a large, complicated selection can take some time. And if it’s an area that needs repeated tweaks, such as for color balancing or lightness, it’s helpful if one doesn’t have to keep going back and reselecting thirty or forty bits and pieces. There’s a feature in Photoshop which can help, but seems almost hidden, in the Channels palette/panel.

Normally, we see the composite channel there (showing the full color image) and the individual color channels (RGB or CMYK, or other) giving a view of the components, which can themselves be manipulated if desired. But we can also store selections in there, in a form called an alpha channel.

Channels Palette

The use of the feature is quite straightforward. First, make whatever selection you need to. Remember you can switch from selection tool to selection tool, changing modes as needed, till you get the entire area you need.


Then, at the bottom of the palette, click the Save Selection as Channel button.

Save as Channel

You can easily rename the channel by double-clicking its name—a good idea, since it’s possible to have something like fifty channels in the palette, and even if none of the selections closely resemble each other, it’s less confusing, and very little additional work.


But what’s even more interesting is, you can fine-tune the selection in the channel by using a painting tool. If you activate the selection (alpha) channel by clicking its name, then select, say, the Brush tool, you can use the color white to add to or solidify the selection if you missed a couple of spots, or black to unselect areas you might have selected by accident.

Fine Tuning

And zooming in to do this carefully is a good trick too. Some people find this is easier for making selections, or at least cleaning them up. It does take a little more time, but since precise selection is often important to professional imagery, any technique you can use to improve the precision is a good thing.

How to Use Bitmap Mode in Photoshop

Of all the image modes in Photoshop, perhaps the least used today is Bitmap mode. It’s the “true” black-and-white mode, unlike Grayscale, which is what a “black-and-white” photograph really is. And Bitmap actually has a couple of advantages, though making proper use of the mode does take a little understanding.

A color picture first has to be changed to Grayscale mode to drop color out, and then one can convert to Bitmap. (If the user knows the image is going to be put in Bitmap, it’s advisable to keep it fairly high-resolution; since Bitmap only has two colors—black and white—compensating with more pixels helps keep the image from looking too grainy.) There are a couple of other things one can do to make a grayscale image better, but that’s another story.

Color Picture Convert Arrow Grayscale Picture

When Bitmap mode is called up, another dialog box asks a couple of questions. One is about resolution—usually this should not be changed. If the image already has adequate detail, it should be left alone. Increasing the resolution should be done beforehand, if at all.

Bitmap Dialog Box

The other is “Method”—how should the pixels be handled? This is the real question, as it affects the final appearance the most.

The first choice, 50% Threshold, looks at all the pixels in the image and applies a straightforward rule: If the brightness of the pixel is 128 or less, it turns black. If 129 or more (the possible range is from 0 to 255), it turns white. But this makes the picture look very harsh and blotchy, as it doesn’t take shading or grays into account.

50% Threshold Picture

The next choice, Pattern Dither, sort of does this. It tries to distribute black and white pixels a little more evenly (in a pattern, hence the name), and can create a fairly “readable” image provided the resolution is not too low. Some people consider this to look artistic, and in some cases the image turns out pretty well.

Pattern Dither Picture

Diffusion Dither, though, is a better choice for a result which looks more realistic. “Diffusion” means the pixels are scattered semi-randomly, the same way perfume gradually drifts through the air. “Dither” is defined as “The use of dot patterns to approximate colors not available in the palette.” (Wiktionary)

Diffusion Dither Picture

Halftone Screen is another matter, and produces something else again. Half-toning was used for a long time in producing newspaper photographs such that they looked like “true” grayscale images. What half-toning really does is to remake the image as a regularly-spaced pattern of dots of varying size to simulate grayscale, rather than bunching them together to make darker (or lighter, by spreading them apart) grays. How often this comes up today depends on what the print medium is going to be, but it’s not as useful for onscreen or conventional computer printing these days since half-tone techniques were developed partly to allow for rapid mass printing (newspapers, magazines) with relatively low-res, high-speed (therefore less expensive) processes.

Halftone Dialog Box Halftone Picture

The advantages to Bitmap mode are, first, it brings the file size WAY down, sometimes only a couple percent as big as the color version, or even less. (Only two colors, after all.) Ditto the grayscale version. Second, though it won’t have color, the detail and simulated gray shades can look almost as good as a true grayscale image, and take a lot less time to print. And finally, it’ll look decent on any printer, color or no. Picking the method is the only thing one usually has to worry about, and often the choice simply depends on the desired final use.

Feathering a Selection in Photoshop

One of the must-have skills in Photoshop is the ability to make precise selections, since we indicate which parts of the picture we want to work with this way, and there are a number of techniques to do it. But equally important is the ability to feather, or “fuzz” the edge of the selection. Very few normal photographs are going to be so supremely razor-sharp in focus that a selection needs to be also, and even those that are will often not be super-high-resolution (that is, naturally a tiny bit fuzzy) anyway. Aside from any specific effects you want to achieve….

Open File

Once you open your file, you can start by making a rough selection with any tool—say, the Magic Wand or the Lasso. Then, either hold down the Shift key or select the Add to Selection button in the tool’s Options at top, and click or drag as needed to add the bits you want. (I admit I like the Magic Wand for a good many items, as it lets me make color-based selections quickly.)

Selection Made

Then, having made and fine-tuned the selection, we go to the Select menu, slide down to Modify, and click Feather in the submenu.

Select Menu Modify Feather Dialog Box

The number of pixels of feather you want to use will depend on a few things: The resolution of the image (the higher the res, the lower the feather number), the area you selected (larger selection=lower feather, generally), and what you want to do with the selected area (you tell me 🙂 ). For things like tinting, a lower number is usually better as you don’t want it to look like smeared paint. For blending effects like softening filters or other non-color-related ideas, a higher number would be good to soften the edge (romantic-portrait-type photos, for example).

It’s a simple effect, but a powerful one. And it doesn’t take too much practice to get a nice result. Besides, you can make a copy of the layer in question and play around with it till you get a good combination, then apply to the final draft.

Clone Stamping in Photoshop

The use of the Clone Stamp tool, in itself, is not hard to understand. It allows the user to “borrow” or copy a piece of picture from one spot to “clone” or paste elsewhere. But some people use it like a paintbrush, and clone big swaths of picture from one place to another—which looks obviously like cloning, or stereotypical “Photoshopping”.

So is there a trick to using the tool and not making things look visibly cloned? Actually, there is. It’s a little bit more work, but it’s not really difficult. It does depend somewhat on the picture and your goals.


Once you open the picture, examine it to see where you want to borrow from and copy to. (Here I use the example of making an elephant disappear—it’d be a cool stage illusion….)

Luckily, the surroundings lend themselves to the effect I’m going for. I start by setting the size of the tool, not too big, and the hardness (how sharp the edge is) to zero, so it blends nicely.

Clone Stamp Tool Settings

I do an Alt-click where I want to borrow from, then use a few clicks to “paste” from that area over the image of the elephant. Alt-click again, from a slightly different area, a few more regular clicks on the elephant, and continue on, each time starting with a different patch, till the modification is done.

Clone Stamping

The real trick—the main point of this kind of exercise—is to borrow a little from here, a little from there. Create the cloned area as a sort of patchwork, to get a somewhat random choice of bits from the “copy” area so the “paste” area actually consists of many small ones, chosen from no one particular part.

Mostly Gone

There is also some technique to clone stamping like this—some study of the image to determine which parts to borrow from and clone to, exactly how big the tool area should be based on the image and its resolution, and so on. But the main thing is to do a little at a time, and not too much from any one area. Nature is usually pretty random when she paints, so following her precept can give you a natural-looking result.

RGB vs CMYK–Which should I use?

Having taught and worked with Adobe Photoshop for many years, I get asked many basic but good questions about it by new users. One has to do with color, and the color systems we use in a picture. There are several that Photoshop can use, but the two most common are RGB and CMYK. The question: Which is best?

The fundamental difference between the two is, RGB is meant for use on screen, and CMYK for print. The terms used to describe how they work are “additive primaries” and “subtractive primaries”, which refer to how these systems show white. For RGB, imagine standing in a dark room with a white wall. Take three flashlights, with color filters (red, green, and blue) and shine them on the wall. Where the three colors overlap, they seem to make white light (the opposite of what a prism does with white light—see Pink Floyd’s album THE DARK SIDE OF THE MOON, specifically the cover—and the back cover.) So the three additive primaries add up to white.

Additive Primaries

With CMYK, the example is even simpler—a piece of paper going through a color printer may have areas where no ink or toner has landed. And the color of the paper we usually use? White, of course. So when the colors are held back, or subtracted, from a spot on the paper, that spot stays (or is) white.

Subtractive Primaries

But we need black ink too, because the dyes or pigments only sorta make black, and a normal eye can see this. So the printing folks added it for completeness.

In doing this, though, we get a couple of problems, and it takes a little thought to get round them. First, because the RGB system normally involves a screen that illuminates itself, it can show more subtle shadings of color than a piece of printed paper (which, unless you’re using radioactive inks—shame on you!—does NOT glow in the dark). The term Photoshoppers and graphic artists use is the “gamut”, which is wider/larger for RGB than for CMYK (where the ink/toner can smear a tiny bit and mess up the shadings). And a printout can’t show all the shades that a screen can. By definition, therefore, printout will always look a little less intense (“saturated”) than onscreen images.

Color Gamuts

And second, any Photoshop image saved in the CMYK system will use 33% more space on disk, regardless, than if saved with RGB. Why? Because the number of color “channels”, how many kinds of color there are in the picture at minimum, is three for RGB, and four with CMYK.

So which should we use?

If you’re scanning in a photograph you want to clean up and reprint (say, from the early 1900s), CMYK will work better, because what you’ll see on screen is what will come out of the printer. But if you’re scanning for archival purposes, scan in CMYK if possible to get a realistic version of the image, THEN change to RGB. Since RGB’s gamut is wider than CMYK’s you won’t lose any subtleties or shadings, but you will get a smaller file on disk. And you can scan in CMYK, do all your work, then save in RGB for later. Best of both worlds. Just have to make sure we understand the tradeoff of size, gamut, and storability.

Free Adobe Photoshop Tutorials

trainingThere are tons of great, free resources on the web for learning a particular piece of software.  At SkillForge, we believe in sharing as many of these resources as possible. Why would a company in the training business want to share training that is free? Well, we know that sometimes you can find a quick answer to your problem by reading a blog post, or going through a quick tutorial. And sometimes, you really need the assistance of an instructor to answer questions or to quickly immerse you in a new product or language. (more…)