|
||
 |
||
|
| Photo-Editing Tips |
 Email This
 Print ThisView My Personal Library |
|
Featured Articles March 2003 • Vol.14 Issue 3 Page(s) 68-71 in print issue |
Photo-Editing Tips Tricks That’ll Enhance Your Image | ||
|
The proliferation of digital cameras and powerful yet inexpensive scanners is giving computer-savvy users much more control over their pictures than they had in the past. The days of filling up a roll of film and impatiently waiting to see how your pictures turned out are quickly fading. Now, it's possible to preview shots directly on the camera, get rid of the ones you don't want on the fly to free up space, and immediately import all your images into your computer for printing. The problem is that all this new technology hasn't necessarily made people better photographers. Shots still come out fuzzy, colors often don't look quite right, images get overexposed or underexposed, and digital data introduces anomalies that we never had to deal with in the past. Fortunately, inexpensive image editors let you correct and enhance images using the same tools professionals have relied on for years. You can read our comparative review, "Recipes For A Perfect Picture," in this issue to find an editor that's right for you; most of the tools and tips in this article apply to any of the popular image editors available today. When you capture images with a scanner or digital camera, the picture is stored as a series of pixels (picture elements). Pixels are individual dots that contain unique color information; the number of pixels in an image defines its resolution. For example, an image with a resolution of 800 x 600 has 800 columns of vertical pixels and 600 rows of horizontal pixels, for a total of 480,000 pixels. The higher the resolution, the more detail an image can have. Think of graph paper: If you draw a large circle on a sheet of graph paper with 100 x 100 squares, it looks coarse with curves that resemble stair steps. Now, draw a same-sized circle on a sheet of graph paper that's the same size, but this sheet has 1,000 x 1,000 squares. Up close, you can still see the stair-step effect, but from a normal distance, your image looks like a circle because there's more detail available to create the curves. Image editors are designed to manipulate pixels and address them individually or in groups. Each pixel contains details about its color, and image editors can modify that color data directly or use mathematical algorithms to apply special effects. They can stretch or shrink a picture, make an image look like an aged photograph, or simply adjust its contrast and color to make a picture look better. The possibilities are endless, and most editors aimed at consumers have dozens of tools that automate the majority of common tasks. The problem is that anytime you change the pixels in an image, you may create unwanted digital artifacts. These artifacts include jagged lines on the edges of objects (called aliasing), blocks of color where there were smooth gradations, and stray pixels that don't blend in with surrounding pixels. There's no way to completely eliminate these effects, but by using the proper tools, you can dramatically reduce the number of artifacts in your edited images.  Compression Many common image formats, such as JPEG (Joint Photographic Experts Group, also called JPG), use compression to dramatically reduce image files while reducing quality as little as possible. There are two basic types of compression, lossy and nonlossy. Lossy compression discards certain image data entirely to achieve better compression ratios and therefore degrades the overall quality of an image. JPEG is a lossy format, and when you save JPEG images, you'll usually see a slider asking how aggressively you want to compress the image. File sizes decrease and image quality degrades as compression levels go up. Nonlossy compression uses less efficient compression algorithms that retain all of the data from the original file when you decompress it. The most popular nonlossy compressed image format is TIFF (Tagged Image File Format, also called TIF). Decompressing an image so you can work with it never affects the quality of the image, but each time you recompress a file using lossy compression, the quality degrades. You can recompress an image as many times as you like using nonlossy compression, and it will never lose any of its original quality. Try to work with nonlossy compressed formats, or even uncompressed formats, such as BMP (bit map), as often as possible when editing; use lossy compression only when you save the final copy of the edited image. Now that you know a little more about the technology behind digital images, we'll move on to the editing tips. Note that various controls and menu items mentioned in this article may have slightly different names depending on the image editor you use. These are the names most commonly used. Resize Pictures Even entry-level digital cameras usually shoot pictures at a native resolution of 1,600 x 1,200 pixels, which captures a lot of detail but is too large to fit on most computer monitors. Image editors make it easy to resize images, but there are several things to keep in mind when doing so.
Resampling is another technique to consider. When you enlarge or reduce an image, the number of pixels in the image must increase or decrease. The image editor uses special algorithms to determine what pixels to cut when the image shrinks and also to determine the colors of pixels that are added to enlarge the image (a process called interpolation). Generally there are three algorithms available: bicubic, bilinear, and nearest neighbor. As a rule, use bicubic interpolation because it samples colors from eight surrounding pixels to determine the color of a pixel being added to the image. Bilinear samples only four surrounding pixels, and nearest neighbor samples one pixel and creates jagged-looking images. Occasionally, bilinear interpolation is better than bicubic because the latter can sometimes make images look too fuzzy. You'll lose detail whether you shrink or enlarge an image. By making an image smaller, there are fewer pixels available to create the image. By making it larger, you lose detail because interpolated pixels make the image look blocky with blotches of color instead of smooth gradations. Always make a copy of the original image before resizing, and if you ever feel a resampling method makes an image too blurry, use your image editor's Unsharp Mask (if available) to reduce the severity of the problem. Despite its name, the Unsharp Mask tool sharpens an image and enhances an object's edges, and it works much better than an editor's Sharpen or Sharpen More tools. Play around with the settings until you find one that makes the image look crisper without introducing artifacts of its own. Let There Be Light
Many novices rely too heavily on an image editor's brightness and contrast controls. The brightness control is usually a slider that makes pictures darker when you slide it left and lighter when you slide it right. At the extreme left, most of the pixels turn completely black. At the extreme right, most turn white. That means as you add brightness, colors become washed out, and as you reduce it, colors become murky. Use this control sparingly and always with the contrast control. You usually adjust contrast with a slider, too, and generally it's more useful than the brightness setting because it can bring out the colors in an image. As you reduce contrast, pixels gradually become drained of color until they turn completely gray when the contrast is at its lowest setting. Conversely, as you increase the contrast, colors become brighter. Your image will begin to lose detail, however, as you max out the contrast. Similar-colored pixels, which are distinct in the original image, will bleed into one another creating an image with large areas of few unique colors. The key when adjusting contrast is restraint. As you increase the contrast, gray begins to disappear from the image as though you were removing gauze that's been covering a print photo. Adjust it too much, however, and suddenly people begin getting big, uniform blobs of color on their foreheads where the light hits them, fabric loses all of its texture, and other details are lost. Try to strike a balance that gives the image the best possible color without sacrificing detail and keep this adjustment to a minimum. Level The Playing Field As useful as the brightness and contrast controls are, if you really want to salvage a faded, poorly lit, or otherwise color-challenged photograph, the image editor's Levels tool is your secret weapon. This tool lets you adjust the tonal range of an image. Most Levels tools come with an automatic setting that makes images look dramatically better, so use Auto Levels whenever possible, but occasionally you'll achieve better results adjusting the setting manually. After opening the Levels tool, an input graph (called a histogram) shows color distribution throughout the image in terms of brightness and darkness. The far left of the graph represents true black (value 0), the far right side of the graph represents pure white (value 255); there are 254 values in between. Usually, you'll see a concentration of colors throughout the middle section of the histogram (represented by hills and peaks), with things tapering off near both ends of the graph. The histogram may also have three triangles you can slide to different positions. The triangle on the far right controls highlights, the one on the far left controls shadows, and the one in the middle controls midtones. There are many ways to manipulate these sliders, but the easiest is to first move the triangle on the far right (Highlights) until it points to the first noticeable rise on the right end of the histogram. With that done, you can move the other triangles to fine-tune the image. In general, move triangles to the right if you want to darken the image and to the left if you want to brighten the image. Advanced Color Correction
We have a photograph of friends standing on a grassy hill (Figure 1) that's badly faded; in it, the grass looks more yellow than green. By using the Hue/Saturation control, which divides the image into even more color groups, including yellow, cyan, and magenta, we were able to select just the yellow channel. Then, using the Hue slider, we changed all the yellow pixels in the image into green pixels, leaving the subjects of the photo untouched (because they contain no yellow pixels). By manipulating the Saturation slider, we made the grass a more vibrant shade of green. This tool allows for many special effects; we easily could turn the grass bright red, for example. Color Balance is another powerful color correction tool that many novices overlook. This tool lets users separately skew the highlights, midtones, and shadows of a picture toward various colors. For example, you can tint the darker, shadowed parts of a picture red, and you can make highlights, such as the sky, cyan. The tool really doesn't make a lot of sense until you try it for yourself, as is the case with many image-editing tools. In our photograph of our friend standing in a field (Figure 2), we show that it's possible to achieve dramatic results with very little work. We removed the red cast from the sky and wheat field by moving the midtones slider toward cyan instead of red, and we augmented the blue sky by shifting the highlights toward cyan, as well. Work With Filters We also applied a sharpen filter to make the picture less blurry. Filters are shortcuts that automate many special effects. You can use filters for everything from blurring or sharpening an image to distorting it or applying effects that make the picture look like a painting. Today's image editors come with dozens of filters, many of which have several presets; you can customize most filters using sliders and other input options. We've already discussed the Unsharp Mask, which "hardens" edges to make them stand out more. Its counterpart is the blur filter, which makes a selection look like it is a bit out of focus but can remove jagged lines from an over-sharp picture and help blend blotchy colors. Many editors come with noise filters that are useful for removing specks of color from a digital camera image or specks of dust from a scanned image. You can use lighting filters to spotlight a certain part of an image or to make it look like a natural light is brightening a scene. There are hundreds of filters, and their effects are cumulative, so don't be afraid to experiment to achieve a unique look. Make Selections The preceding tips are all basic edits for an entire image, but many times, you'll want to edit a portion of a picture. Nearly everyone knows how to use the standard rectangle and oval selection tools, but there are other options, such as the Lasso and Magic Wand you may have, that work better in certain circumstances. Use the Lasso to select freehand shapes or when you need to trace around an object. Click and hold down the left mouse button and draw the shape you want to select. The Magic Wand selects groups of pixels that have the same or similar colors. Usually you can adjust the threshold, telling the wand to select only one particular color or many shades in the same range. The magic wand is useful when you want to select a large patch of relatively uniform color, such as the sky. Many times you'll want to add a portion of the image to the selection, deselect a single portion of the image, or create multiple selections throughout the image. You can do any of these by pressing the SHIFT or CTRL keys in conjunction with the selection tool. If you want to add to a selection without deselecting portions of the image that are already highlighted, press the SHIFT key while making the selection. To subtract from a selection that already exists, hold the CTRL key and select the portion you would like to remove. Fix Scratches & Remove Elements Many computer users like to store old family photos by scanning them into their computers, and aside from having to correct the inevitable color fading problems, many old photos also have scratches, flecks, or other damage that's particularly noticeable when the image is scanned. Some editors have built-in damage repair tools, but if yours doesn't, using an image editor's Cloning or Rubber Stamp tools is easiest. A cloning tool work like a paintbrush that paints an image instead of a solid-colored line. You select a portion of the image you want to clone (generally an area adjacent to the scratch or other flaw) and then when you click the scratch, it is covered by a duplicate of the portion you selected. For example, if there is a white scratch in a blue sky, you can select a portion of the undamaged blue sky and then "paint" a copy of that sky over the scratch so it blends in with the rest of the image. In this example (Figure 3) we've removed a hand from the scene by using the cloning tool to "paint" carpet over the appendage. It helps to clone different parts of the scene every so often so that patterns don't develop that give away the fact that you used a cloning tool, and the tool works best when there is a solid or patterned background to clone. It Doesn't Stop There This article barely scratches the surface of what a modern image-editing application can do, and there are hundreds of add-on products available that integrate with most editors to add even more capabilities. If you have an image editor that is compatible with Photoshop plug-ins and filters, make sure to take a look at our Web-only companion article "Focus On Filters" that reviews some of the best freeware and professional plug-ins available. With some knowledge, a lot of patience, and a little money, anything is possible.  by Tracy Baker
|
|
Home Copyright & Legal Information Privacy Policy Site Map Contact Us