|
|
Any change you make takes information away from the image file. It's a subtractive process. You can make the photo look better, but you will be taking data away. Therefore you always want to do as little as necessary when working on a photo. Less is best.
"Auto" adjustments are making a change based on numbers that are either built-in for what an average change of that type would be, or you can control the parameters of that change (with Photoshop one way would be by setting numbers in the Color Picker). If you don't know what values the editing software is using for "Auto" fixes, be cautious.
Editing software depends on RAM for speed. The more photos you have open in the program, the more RAM you'll need to make things happen at a reasonable speed. If you don't have a gigabyte of RAM in your computer, avoid having more than a few photos open at once.
Put some thought into what format you save your images. If you're starting with a JPEG, then remember that if you make any change and the save, you've just re-compressed that file again. Each compression cycle throws away some information, regardless of the level. Therefore avoid making lots of saves to a JPEG file, and consider perhaps saving the final image as a TIFF (see more on saving and formats later in this section).
Never work on an original file. You'll never have more information in a file than the one that came out of your camera. It's important to always be able to go back to that original if you need to. Therefore always work on a copy.
Remember, unless you're working on a monitor that's been properly calibrated and profiled with a hardware/software package, you can't trust what you're seeing on screen (see the section on Color Management for more on this). This is the single biggest reason people think their camera or printer is not working properly, when in fact it's simply the display. If you don't want to pay the extra money for one of these packages, just remember that what you're seeing on your monitor won't be exactly what's actually there.
|
|
1 - Crop
Might as well do this first. It's especially important if your computer doesn't have a lot of horsepower (RAM and fast processor), as the smaller image size will mean fewer pixels for it to push around.
2 - Tone/Color
Obviously, these are two steps, but I decide which to do first based on which is the bigger problem. If the color's okay but the image needs to be lightened, darkened, I'll do the toning first. If the color is way out of whack, I'll fix it first. Once the larger problem is solved, then I can see better what else truly needs to be done.
3 - Fix digital problems
This can be a broad area, but the idea is to fix any problem that is digital-camera specific. Some cameras are more prone to moiré patterns, so you may need to deal with that. All cameras have at least some digital noise, especially at higher ISO settings, so that can be a concern as well. Understand where your camera's strengths and weaknesses are, and be prepared to deal with the weaknesses.
4 - Save
Once I've done all the work needed to get the image where I like it, I want to save it at that state. I prefer to save it as a TIFF, so that I'm not introducing any compression to the file. Even better, if I've been working in Layers, I'll save it as a layered TIFF, so I can go back and re-work those layers any time I want.
5 - Size for output
Working from the saved TIFF, I can now re-size the image to make a small or large print, or to prep for email or web use.
6 - Sharpen for output
After sizing the image the way I need to, I can now properly sharpen it based on that size and how it will be viewed (print or monitor).
|
|
Top
Let's talk about these different steps, and the different tools available to do them. I'm going to use examples from Photoshop, but many of the tools and techniques are common among other editors as well.
|
|
Top
Levels
More advanced users will want to look for manual control to make these changes. Levels is a tool common to many programs, and generally gives a histogram display (a graphic readout of pixel tone values) with a shadows slider at the left, a midtone slider in the middle and a highlight slider at the far right. Moving those sliders effects changes in those values.
Adobe Photoshop Elements and Adobe Photoshop both offer a
more graphic way of helping you set white and black points
(whitest and blackest pasts of the image) with Levels. With
the Levels dialog open and Preview checked, hold down your
Alt (Windows) or Option (Mac) key and move your highlight
or shadows slider. When you do this with the shadows slider,
you'll first see the image (or at least most of it) go completely
white. As you move the slider to the right, more pixels
will start appearing. This is called Threshold mode, and
Photoshop is showing you which pixels are being "clipped,"
meaning forced to black. You can do the same thing on the
right side of the Histogram, with the highlights slider,
only now the image starts completely black. As you move
the slider to the left, you'll see pixels appear as they're
clipped and forced to white. This is a manual and very interactive
way to set a white and black point for your image.
|
|
Top
Curves
Another tool that offers finer control over making adjustments is called Curves. Curves also starts with highlight and shadow points, but the tool's display is a line stretched from the lower-left corner of a box to the upper right. "Grabbing" the line in the center with your mouse, you're able to adjust midtones while bringing along the three-quarter tones (between midtones and shadows) and the quarter-tones (between midtones and highlights) along to a lesser degree. The real power of Curves comes in the ability to set more than one point, and thus adjust different tonal areas in the same adjustment. Photographers often use Curves to add a bit of contrast to an image, by setting two points and creating an "S" curve. You do that by either pushing up or down on the midpoint, then pushing the quarter or three-quarter tone section the other way. The more you push, the more the curve takes the shape of an "S" and the more contrast applied. Remember, though, that in Photoshop, less is often more - don't get carried away, or you'll create an image that has too much contrast and won't print well.
|
|
Color in Levels/Curves
So far, we've been making adjustments to overall lightness/darkness by adjusting the RGB values together. You'll notice that both Levels and Curves have a pull-down that lets you adjust just one channel, Red for instance, separate of the other two. This is a basic way to do overall color changes.
|
|
Top
History Brush
As you become more adept at working in Adobe Photoshop, you'll learn how to use the History palette and History Brush. The History palette tracks all of the changes you make to your image. It's sometimes referred to as a "multiple-undo," but that doesn't give credit to its real power. The strength of the History Palette comes in using it with the History Brush. In Photoshop's tools palette, under the Brush, you'll see another Brush with a curved arrow wrapped around it. Click that icon to make it active, then click in the square box at the left of any History State above the current one. You've just told Photoshop that you want to paint the pixel values from that previous state of the image into the current state (the image in front of you). Now if you paint on the image with that brush, the pixels in any area you touch will be taken back to that point. You can do this in a more effective way by using the Brush tool's Options bar at the top of the window and setting the opacity to less than 100%. I often set it for about 35%, which means that each time I paint into the current version, I'll get 35% of what it was before. It lets me blend in the change very nicely. Unfortunately, if you're an Adobe Photoshop Elements user, you'll notice you have no History Brush. Don't despair, though, because you have an eve more powerful tool available - Adjustment Layers.
|
|
Top
Adjustment Layers
If you like the History Brush, then you'll love Adjustment Layers. I like to call it "History Brush on steroids." One of the problems with working any digital image is that any change you make takes information away. That gives you even less to work with for the next change you need to make. Also, if you realize two or three steps down the way you made a mistake early on, you have to throw away all the other work and start over. What if you could make changes, but see them as a preview, without actually making the change? What if you could make dozens, or even hundreds of changes, and go back and re-work any of one them without affecting the others? This is what Adjustment Layers does, and why it's such a wonderful tool. It's also common among mid-to-high-level image editors, not just one or two programs. It's a little complicated, but well worth taking the time to learn.
The first step is to find out if your editing software supports Layers, and if so, how to create one. With adobe Photoshop or Adobe Photoshop Elements, you go to the Layer - New Adjustment Layer menu. In there you'll find many of the adjustment tools you're used to seeing. If you create one in this way, a dialog box will pop up asking if you want to name the layer. Since you can create hundreds of layers (and graphic artists often do), naming them can be very handy. For a photographer who normally uses only a handful of Layers, naming them is more trouble than it's worth. In that case, there's another way to create a layer. At the bottom of the Layers palette you'll find a small half-white, half-black circle. Click on that and you get a pull-down with the same adjustments. Choose one - I usually use Levels or Curves, and the tool pops up on the screen. Also notice that you now have an Adjustment Layer active in the Layers palette. Notice that underneath it is the "Background" image. This is your real image data, which will be untouched until the end. Now for the fun stuff.
Let's say you've chosen a Levels adjustment. Go ahead now and make a change. You use the tool the same way you're used to. In fact, go overboard and make too big a change. Now OK it. This is the point where in the past you'd have to undo and start over. Not with Adjustment Layers. There are three ways you can re-work what you've just done:
|
|
That Adjustment Layer is still active (notice it's selected in the Layers palette). See the little black/white circle on the left side of it in the palette? Double-click on that and the Levels dialog pops back up, just the way you left it. Now you can re-work it if you like.
Here's another way to change it. Notice the Opacity slider at the top right in the Layers palette? You can click on that and drag the slider to the left to lessen the effect of that change. I like to call it an "intensity" slider, as it reduces the intensity of the change you just made with that layer.
Lastly, you can make selective changes to the image using that Adjustment layer. This is the tricky one. Photoshop will let you erase the change you just made. With that layer active, select the Eraser tool in the tools palette. The tricky part here is making sure that you've got your foreground color set to white and the background to black. That's done with the little white and black square near the bottom of the Tool palette. The tiny white/black squares at the lower left sets the larger one to white and black, and the two-headed arrow at the upper left switches black to foreground or vice versa. With white as the foreground color, you can now use the Eraser tool to erase what you did in that Adjustment Layer. I prefer using a soft-edged brush, and the bracket keys will grow or shrink the bursh size as needed. To make the change gradually, use the Eraser options tool at the top of the window to set the Opacity to less than 100%. I like to use 25% or 35%. If you do this, then each time you click and drag, any pixel touched will be taken back that percentage from the change in that layer. If you're using 25%, the first click-and-drag would take you back 25%, the second over that same area 50%, and so on. This is an easy way to feather the change.
|
|
You'll be amazed at how much you can do with Adjustment Layers, and how easy it is once you grasp the different ways of using them. When you've got an image just the way you want it, you then have a couple of choices about how to save:
|
|
The best way is to save the file with the layers. You can do that by saving it as a PSD (Photoshop Document) or as a TIFF. With either one the default is to save it with layers. This is great, because when you open the file at a later time, you'll have all those layers you worked back again, ready to be re-worked if necessay. The only real downside to this is that the file will be BIG, as those layers add real size to the file.
If you can't save with layers, don't want to, or need a JPEG, then you'll need to Flatten the image. You can do this from within the Layers palette. Look at the upper left corner of the palette and you'll see a small triangle. Click and hold on it nad you'll get a pull-down, with a choice of Flatten Image. This will collapse all the layers into the background image and finalize those changes. Now you can have a standard TIFF or save it as a JPEG.
|
|
Use the Edit pull-down to select the color you want to work on.
When you do that, the eyedropper at the bottom right becomes active. That's Photoshop's way of saying, "please show me exactly what color you want to work." Use it to click on the color you're trying to adjust. That allows you to target the specific color.
Now you can use the sliders to fix that color, and restrict your changes to just that color. Be sure and turn the Preview on and off to make sure you didn't unintentionally change another part of the image. If so, use either the Eraser tool with Adjustment Layers or the History Brush with the History palette to fix it.
|
|
Choose Channel Mixer, or, better yet, create a Channel Mixer adjustment layer.
Click the "Monochrome" button at the bottom left. That still leaves you with your color channels, but the output channel is now Gray.
Adjusting the Red and Blue sliders allows you to change the overall tone and contrast. I tend to leave the Blue alone, and the Constant can create massive tonal shifts.
Once you've got the look you like, you can click the OK button. Now you've still got an RGB image, but it is being displayed in grayscale. At this point you can use Image - Mode - Grayscale to reduce the size of the file but keep the same look.
|
|
There's another technique I like to use sometimes after converting an image to black-and-white this way. It's a nice way to add a little (or a lot, personal choice) of color to a black-and-white photo. To do this, don't take the final step of making the Channel Mixer image grayscale through the Image - Node - Grayscale command. Instead, choose Layer - New Fill Layer - Solid Color. Now choose Color Burn from the Mode pull-down, and click OK.
At this point the Color Picker will pop up. Choose a color from the vertical rainbow strip in the middle, then click on the density of the color in the large "Pick a Solid Color" box. Since this is a layer, you can use the Opacity slider in the Layers palette to reduce the intensity of the color. It's a nice, easy way to get a duotone-look.
|
|
Open the image with the color cast problem.
Open Levels, or better yet, create a Levels Adjustment Layer.
Double-click the eyedropper at the bottom right of the Levels dialog that is the neutral you're going to use (white, neutral gray or black eyedropper).
The Color Picker will pop up. Click on the spot in your image that should be neutral (free of color), and then don't touch the mouse again (you want to use that same spot again soon). You'll see the Color Picker change to reflect the color you just clicked on.
Using the Tab key to move through the fields, reset the two lower RGB values to the same number as the highest (i.e., if the values read R167, G142, B98, then set all three to 167).
Use the Return key to "OK" the changes.
The eyedropper will still be active, so now click that same spot again, and Photoshop will take those numbers and change not only the spot you clicked, but the entire photo based on the numbers you plugged in. Voila!
Now when you close the Levels box you'll be asked if you want to "Save the new target colors as defaults?" Choose the "No" button, which will be the default.
|
|
There are a couple of keys to doing this effectively. You want to find a neutral in the image that has some density (not overexposed or nearly so). A neutral that has some shadow will often show the color cast more effectively than one without. Finally, if you do this changing two of the RGB values as above, you may find this trick unintentionally lightens your image to the point where you lose highlight detail. If that happens, try the routine again, but this time reset the "a" and "b" numbers in Lab Color to 0 (zero). That will clear the cast with little change to the density. The RGB method is a little cleaner, but the Lab way will get the job done as well.
|
|
Go to Image - Mode and change the image to Lab Color.
Now move to the Channels palette and touch on the "b" channel to make it active.
Go to Filter - Noise - Dust and Scratches. Move the Radius until you blur the chunky look. Click OK. The "b" channel is where the majority of your noise will be.
Do the same to the "a" channel, but to a lesser degree.
Now go back to Image - Mode, and change the image back to RGB.
|
|
Top
RAW image editors
With high-end compact cameras and digital SLR's, you'll often have the chance to shoot the camera in RAW format. That means that the camera records the data from the sensor with little processing. The advantage is you can decide later, at the computer, what white balance to render the images at, the amount of noise removal to apply, sharpening, type of tonal curve and other parameters for the processing of the image. The downside to RAW files is that they're larger than JPEG (but smaller than TIFF), and you need special software to do that processing with.
Every manufacturer has their own software they sell or include with the camera (Nikon Capture is one example), and there are third-party programs (such as Phase One's popular C1 Pro) as well as the Adobe Camera Raw plugin for Photoshop 7, which is built into Photoshop CS. The advantage to a manufacturer's software is that they understand the strengths and weaknesses of the camera's sensor and can compensate for that in their software. The third-party and Adobe software is often faster, with a better interface. The advantage to the photographer is that you have several different options in how to process your RAW images, a choice that can give your images different "looks." Try the different programs and find the one that suits you best. Of particular interest to professional photographers will be the batch processing capabilities that most of these programs have. These functions allow you to automate the processing of whole folders of images.
Top
|
|
|
Digital
Cameras
Camera
Features
Photo
Techniques
Computers
Downloading
Back
Up
Color
Management
Image
Browsers
Editing
Printing/E-mail
|