Print17 slides p6

Editor's Notes

• #2: With the increasing use of digital cameras in the hands of non-professionals, many images used in professional design jobs will require at least some assistance. This project highlights two of the most common tasks that a Photoshop artist might encounter in a professional environment: retouching and color correction. The image in Stage 1 of this project requires a number of common retouching tasks associated with removing damage caused by improper storage. There is an entire industry built around restoring old photos, so you might find these skills useful either as part of the job — or even as the entire job. We selected the various images in Stage 2 and Stage 3 of this project to illustrate a number of common color correction issues, from basic tools for adjusting overall contrast to more sophisticated options for correcting even a single color channel. These skills will be valuable for many types of graphic design projects, whether it is simply enhancing the overall appearance of a photo or creating a specific color match for retail product photography. It is important to emphasize that correcting color for a print project depends largely on the gamut that is available in the defined output profile. Stage 4 of this project focuses on this issue.
• #3: In traditional photography, grain was the result of small particles in the film that are visible in the final print. Grain can be particularly evident in large areas of solid color (especially lighter colors, such as a face or a sky). Removing grain is easily accomplished through a sequence of blurring and sharpening techniques. Basically, you slightly blur the entire image to remove the visible grain, then sharpen the image to restore the details that were blurred in the first step of the process. Be stingy with how much blurring and sharpening you apply. If you blur the image too much, Photoshop simply will not be able to restore the detail that was present before blurring. Use only the amount of blurring that is absolutely necessary to remove the noise or grain. On the other hand, sharpening can restore a lot of detail; however, too much sharpening can create halos at high-contrast edges that are possibly more offensive than the defect you wanted to remove in the first place.
• #4: Photoshop includes a number of different Blur filters, many of which are more useful for artistic effects. The Gaussian Blur filter, used to retouch the photo in this project, also has a common functional purpose of removing noise and grain from an image. A very small-radius Gaussian blur can help to correct a number of small defects in an image without causing significant damage to areas of detail. Three of the Sharpen filters —Sharpen, Sharpen More, and Sharpen Edges — allow no user control over the amount of sharpening that is applied. After applying a Gaussian blur, sharpening helps to restore some of the detail in the image. The Smart Sharpen filter provides fairly extensive control over the sharpening process. To avoid over blurring or over sharpening, some Photoshop artists actually prefer to use the blur/sharpen sequence multiple times — using very small blur (0.5-pixel radius or less) each time — to remove this type of defect.
• #5: As with so many different Photoshop tasks, there are a number of different ways to remove objects from an image. These four tools each provide different types of retouching capabilities. It is a good idea to determine what you want to accomplish before you decide which tool will best work for that specific goal. The Spot Healing Brush tool simply blends surrounding pixels in the area where you click. It is useful for removing small spots in areas of relatively solid color, but not usually appropriate for removing damage in areas with sharp detail. The Healing Brush tool blends the color of pixels where you click with the color of pixels in the user-defined source. This tool has a similar function to the Spot Healing Brush tool, but you can define the color of pixels that will be blended to heal the damage where you click. The Clone tool “paints” pixels from a defined source area — including an area of another image — onto another area. Because you can manually define the source pixels, this tool can be useful for healing damage along defined edges and other areas of detail. The Patch tool copies entire sets of pixels in a selected area. It can be useful for removing damage in detailed areas because the application attempts to blend the edges of the original and the replacement pixels.
• #6: As with all tools, check the Options bar when you use a healing tool. The Spot Healing Brush, Healing Brush, and Clone tools are all based on a defined brush, which you can change using the Brush Preset picker. For the Spot Healing Brush and Healing Brush tools, Photoshop automatically blends the edges of the brush diameter to avoid visible “spots” of healing. The brush hardness value is not as important a consideration as it is with the Clone tool. More importantly, you should use the smallest brush possible to repair the damage. For the Healing Brush tool, the Aligned option can be tricky. If the option is turned off, the source pixels revert back to the original source every time you release the mouse button. (See Page 339 for a detailed explanation of the concept of alignment.)
• #7: With the Clone tool, changing the hardness of the brush edge determines exactly what will be cloned. For example, a Hardness of 100% means the clone brush might result in a hard-edge circle (based on the brush size) if the pixels where you click are even slightly different colored than the defined source pixels. Reducing the hardness creates a soft-edge brush, which allows the cloned pixels to better blend into the surrounding area. The Clone Source panel (explained on Page 343) can be useful for controlling exactly how the source pixels will be cloned, as well as previewing the potential result of the clone using a transparent overlay. Tip: When using the Healing Brush or Clone tool, most professionals frequently resample the source when healing or cloning large areas of an image. This helps to minimize obvious patterns in the repaired areas.
• #8: The Content-Aware Fill option is sometimes called “magic” fill. The Adobe demo used to show a buffalo being removed from the middle of a grassy field, almost like magic. The application generates fill pixels based on the surrounding areas, so the random grass nicely filled in the spot where the buffalo used to be. It is important to know, however, that this feature does not work well when trying to fill areas that include well-defined edges. In the example of this project, you are using the option to fill in several scratched areas in a photo. You first clone the sharp edge of the man’s shoulder, then use Content-Aware Fill to fix the areas around the shoulder line without selecting any part of the high-contrast edge. If you simply selected the entire scratch, including the shoulder, the application would have a very difficult time calculating the new pixels along the jacket edge; the results would not be as good as what you accomplished by first fixing the edge with the Clone Stamp tool.
• #9: Unless you are working with a professional photographer for all of your work, you will almost certainly find images that were captured in less-than-ideal lighting. Stage 2 of this project explores a number of adjustments that can improve these images. Keep in mind that there is no substitute for good original photography. The adage “garbage in, garbage out” might be just a bit harsh, because Photoshop can help bad images — but the application isn’t magic; it can’t make something out of nothing. Before you start trying to correct lighting — or any color issues, really — you should understand the basic terms related to lighting. •Highlights are the lightest area of detail in an image. •Specular highlights are areas of pure white, where there is no discernible detail. Examples include a lit bulb, or a sun glare off a lake surface. •Midtones are the middle range of tones, halfway between the highlights and shadows. Midtones are also referred to as “gamma”. •Shadows are the darkest areas of detail in an image. You might also hear the terms quartertones, which are halfway between highlights and midtones; and three-quartertones, which are halfway between the midtones and shadows.
• #10: The Brightness/Contrast adjustment is a basic correction tool that can be useful for overall image problems; it does not provide extensive control over different image details. Brightness is the overall apparent luminance of the image. Increasing brightness lightens all tones in the image; decreasing brightness darkens the entire image. Contrast can be defined as the overall tonal variation in an image. Increasing the contrast increases detail throughout the image; decreasing contrast creates a less-detailed, often “muddy” appearance (especially in the midtone range). Saturation, or the intensity of color, is closely linked to contrast. Undersaturation adds a gray cast to colors and reduces the apparent contrast; oversaturation results in a too-bright, almost neon appearance of colors.
• #11: Before you make adjustments in the Levels dialog box, you should understand the concept of tonal range. Basically, there are a specific number of tones available in each color model; image quality suffers when an image does not take advantage of all the available tones. The histogram is a visual depiction of an image’s tonal range; it can be viewed in both the Histogram panel and in the Levels dialog box. Peaks in the histogram also show concentrations of tones. For example, the chef image in this project shows peaks near the highlight and shadow point. The valley in the center of the histogram indicates that the image does not have a lot of midtone pixels. This distribution gives you an idea of which areas might benefit from an extension of the tonal range.
• #12: Using the Levels dialog box, you should first define the highlight and shadow point of an image. This allows the application to extend the image tones across the entire available tonal range. Increasing the tonal range allows more variation, which increases image detail/contrast. Think of it like this: There are 256 possible tones in the RGB color mode. Say an image histogram shows an image is only taking advantage of 200 of those tones (its original tonal range). If you extend the tonal range to the full 256 available tones, you allow 56 more tones to depict different details in the image. After defining the highlight/shadow points, you can also drag the Gamma Input slider to extend the tonal range of either the highlight or shadow half of the tonal range. Again, more available tones means more contrast in that range of the image. Keep in mind, however, that increasing the tonal range in one half of the image automatically decreases the tonal range in the other half. In other words, increasing the highlight range also decreases the shadow range and decreases the contrast in the shadow areas. As with so many other issues, you have to consider what you want to accomplish, and where you are willing to make compromises. The shadow areas in the chef image are not as important as the highlights, so increasing the highlight range is an acceptable trade-off.
• #13: Although the exposure adjustment is intended for HDR images, it can also be used to correct overall under- or overexposure problems. The Exposure slider affects the overall luminance of the image. The Offset slider affects the highlight and shadow points; the white point is not changed. The Gamma Correction slider adjusts the image midtones.
• #14: Stage 3 of this project focuses on color correction, which can be one of the most intimidating aspects of Photoshop. Although you don’t need to be a color scientist to understand the process, you do need a basic grasp of color terminology terms to understand the concepts that are vital to color correction. The information on Pages 184-186 is very brief, but very important. In the RGB color model, equal parts of each primary color creates a neutral gray. This is what we mean by the term gray balance — when all of the primary colors are equal, or balanced. If any of the primary colors has a stronger presence than the others, areas that should be neutral will show a marked color cast of that primary. Any color cast will be most evident in neutral areas, but will also affect the entire image. Correcting any problems in gray balance will go a long way toward correcting the overall color in an image. Reading the Info panel is the first step to finding color problems. The Color Sampler tool places markers, and the Info panel shows the color makeup of the pixel under each marker. When you place a marker in a known neutral area, you can quickly see what, if anything, needs to be corrected to achieve gray balance.
• #15: Using the principle of complementary colors, you can correct gray balance problems by either subtracting the predominant primary or adding its complement. The Color Balance adjustment provides a basic interface for exactly this type of correction. The three sliders show the three complementary scales: cyan/red, magenta/green, and yellow/blue. This adjustment is useful for a quick fix, but is not as precise as the Curves dialog box.
• #16: The Curves dialog box is another tool that suffers from the intimidation factor. It is one of the most powerful options for precise control of image color, but it should not be used without a basic understanding of the issues discussed up to this point. You can adjust curves for the entire composite image, or for individual color channels. The Curves dialog box shows the same histogram that you would see in the Histogram panel or the Levels dialog box. If the histogram does not take up the entire tonal range, you can adjust the Input sliders below the graph to extend the tonal range. Clicking the graph adds a point to the curve. When a specific point is selected, you can define specific input and output values for that point. Again, aim for neutral grays; that can go a long way toward correcting overall color problems. Keep in mind that the curve is a line; adjusting one area of the curve has an inverse affect on the opposite area of the curve. A steeper curve indicates increased tonal range, which increases contrast in that area. It follows, then, that flattening the curve decreases the contrast in that area. For example, steepening the curve in the three-quartertone area causes the curve to flatten in the quartertone area. Contrast is increased in the three-quartertones, but decreased in the quartertones.
• #17: If you are designing for print, the conversion from RGB to CMYK can cause significant problems when the job is output. This is the focus of Stage 4 of this project. At the very least, you should understand the difference between RGB and CMYK, and how the difference in gamut can cause problems. More specifically, you should understand that CMYK is the ultimate color model for most (if not all) commercial print design.
• #18: Color management, explained on pages 363–365, has the goal of consistent color output across various output devices in the design process. We tried to keep this information brief, but it is a crucial bit of knowledge for any designer who works in the print design world. It is important to understand that the ultimate goal of color management is to best represent the ultimate output on various intermediate devices (monitor, proofing devices, etc.). This requires calibrated devices throughout the process; if you are using uncalibrated equipment, color management is still a best-guess scenario. Color management is intimidating to many designers, especially new users. Remind students that they don’t need to be afraid of it. The most important thing is to be aware of the issues related to different color models, and translating color from one model to another. Most importantly, students should understand that many RGB colors are not available in the CMYK gamut.
• #19: In Photoshop, applied color management means defining working profiles for the various color models. •The working RGB profile tells Photoshop how you are seeing the color you work with on the canvas. •The working CMYK profile tells Photoshop how the ultimate output process will create color. This information is vital for Photoshop to effectively manage the conversion from one space to another, whether temporarily displaying it on your monitor or when converting an RGB file to CMYK. An embedded image profile tells Photoshop how the capturing device defined color. As a general rule, you should not maintain embedded profiles, so Photoshop can correctly translate the colors in the image to the colors in your defined working profiles. Two options in the View menu apply color management in Photoshop’s design environment, allowing you to preview the conversion from a file’s color space to the working output profile: •Gamut Warning shows areas of an image that are outside the defined output profile’s gamut. •Proof Colors shows you how colors (including out-of-gamut colors) will appear once converted to the defined output profile.
• #20: Remember: The RGB and CMYK color models have very different gamuts. Most (if not all) of your source images will use the RGB color model. To print commercially, those images must be converted to the CMYK space. Because the CMYK gamut is so much smaller than the RGB space, some color shift will almost always occur during the conversion process. Some people ask, “So why not convert to CMYK and then do all of your color correction?” By taking advantage of the greater tonal range in the RGB space, you can achieve better contrast throughout the image. You can then determine where problems might occur during the color conversion, and correct only what is necessary. If you convert to CMYK first, you can lose detail in important areas before you ever begin the correction process. Photoshop’s Proof Colors and Gamut Warning options (in the View menu) are both helpful here. Keep in mind, however, that both of these rely on accurate color profiles and calibration. Without those two things, you should never rely on your monitor to know what an image will look like when it is printed.
• #21: The physical limitations of the printing process require a few different adjustments when you prepare an image for commercial print output. First, you should understand the ideas of the minimum printable dot and maximum printable dot (see Page 370). Basically, you have to identify the lightest highlight and the darkest shadow in an image, and tell Photoshop what ink components will be used to output those areas. Remember: Highlights and shadows are defined as areas of detail; they do not refer to solid white or black areas with no discernible detail. Total ink coverage requirements restrict the color makeup of the maximum printable dot. You simply can’t put down 100% of all four primary inks to create the darkest shadow; the paper would be oversaturated and tear (among other problems). You can use the Curves dialog box to define the target highlight and shadow colors, and then identify those areas in the image by clicking with the associated eyedroppers. After making these two corrections, you can easily see any other areas that should be corrected if you have the Gamut Warning turned on.
• #22: After you finish correcting an image based on the known CMYK parameters, you can safely convert the image color model. The Image>Mode submenu provides all of the available options. When you choose one of these options, the image is converted based on the defined working profile for the color mode you select. If you are not sure of the current working profiles, you can choose Edit>Color Settings to review or change them. Say the defined working CMYK profile is U.S. Sheetfed Coated v2. Choosing Image>Mode>CMYK converts the colors in the image based on information in the U.S. Sheetfed Coated v2 profile. You can convert the image mode to something other than the working profile by choosing Edit>Convert to Profile. Of course, keep in mind that the Proof Colors and Gamut Warning previews are based on the defined working profile by default. If you preview and correct to one profile and then convert to another, you can still get some unexpected results. If you want to preview a profile that is not the active working profile, choose View>Proof Setup>Custom, and then choose a different profile in the Device to Simulate menu.
• #23: Although not technically a required part of this project, it would be an obvious omission not to include a brief mention of the issues related to converting a color image to grayscale. This topic is discussed in detail on Page 375. There are many times when you need (or just want) to convert a full-color image to grayscale. As with any other mode conversion, you could simply choose the appropriate menu command and let the software perform the conversion. However, this is uncontrolled, and — in this case — is really not the best method for creating a grayscale image. Simply changing the color mode flattens the three component channels into a single Black channel, using the average pixel value from the original channels. Contrast — and thus detail — are almost inevitably lost. The Desaturate adjustment produces slightly better results, averaging the individual channel values and then using the same average value on each channel. The Black & White adjustment is the best possible approach because you can carefully control the data that will be used on each color channel. Both the Desaturate and Black & White adjustments maintain three component color channels; to create a “real” grayscale image, you then have to choose Image>Mode>Grayscale.
• #24: HDR (high dynamic range) images are becoming more common, especially as more consumer-lever cameras make it easier to capture the multiple exposures required for this type of image. HDR images often have a surreal look because we are simply not used to seeing photos with this much color definition. As the the process gains wider acceptance in the mainstream market, HDR might become the standard instead of the exception. (Think about the difference in quality from an old home movie — if you have ever seen one — compared to a digital video.) Stage 5 of this project explores the HDR capabilities that are built into the Photoshop interface. You can combine multiple exposures of the same scene to maximize detail in the composite image. Keep in mind that an HDR image is 32-bit. If you are going to use an HDR image for print, you will eventually have to convert it to 8-bit. You can do this directly in the Mode menu of the Merge to HDR Pro dialog box, or by choosing Image>Mode>8 Bits/Channel in the primary Photoshop interface. When you choose the 8-bit option (regardless of which method you use), you can experiment with the various settings to adjust the exact details in the image. The ultimate result is largely subjective.