Read 1571691561.pdf text version

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 1 of 27

[Figures are not included in this sample chapter]

Adobe Photoshop 5 How-To -3Scanning

How do I...

3.1 Get the best image information in my scans? 3.2 Compensate for dust in my scans? 3.3 Select a reflectivity for my original before scanning? 3.4 Pick a color setting for scans? 3.5 Choose scanning dpi? 3.6 Choose a level of sharpness for scanning? 3.7 Choose a level of descreening for scanning? 3.8 Choose black and white points? 3.9 Scan a 3D object? 3.10 Scan a translucent object? 3.11 Scan an oversized image? One of the most common ways to incorporate an image into the computer is through scanning. Though scanning is not directly a part of Photoshop, the ability to scan pictures, and understand how to do it, is. Getting good, consistent results enables you to use a much broader scope of premade materials (notice I did not say "pictures") to enhance your work in Photoshop. Too often people either rely on the presets of a scanning program or just press the auto setting and hope for the best. A person who knows how to scan will start with a better image, and at that point, time--if not money and frustration--are saved. Certainly you can use the tools in Photoshop to fix whatever goes wrong. The purpose here is not to get the most amazing scan in the world, but the best one possible with your available equipment (hardware and software) and one closest to your purposes. If you don't own a scanner, one will probably be accessible at a local self-service copy place and perhaps in a local library if you are lucky. Collect some interesting things to scan and be sure you bring compatible removable storage if you are doing the scanning elsewhere so you can bring the images home and get to work after they are scanned. For the most flexibility with scan size, use larger format removables. ZIP disks are common and a convenient size to hold even large scans. Most desktop scanners have a maximum scan size of 8x14, and a ZIP can hold a file that large at a bit more than 450dpi in CMYK without compression. You can scan three-dimensional objects as well as flat items using the scanner as a somewhat limited digital camera. It can be good for creating product shots or capturing the likeness of any number of items that you want to incorporate into your Photoshop images. From paper clips to camera gears, statuettes, and fabrics, a scanner can help provide raw and finished materials that you can use in developing images, textures, patterns, and backgrounds to use in any way you can imagine.

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 2 of 27

Most of the information in this chapter applies to consumer-grade flatbed scanners. The sections in this chapter help you make better scanning choices. TIP For practice in scanning, it might be a good idea to follow the steps as closely as you can in section 3.1; then refine your scans and scanning ability later by progressing through the stages in the chapter. This way, you can see the changes and improvements in your scans and learn what works to make them better. 3.1 Get the Best Image Information in My Scans A scanner can get different image information from the same object depending on how it is used. Getting a better scan is the difference between fighting with an image that doesn't have the right information and simply using the images you need after you scan them. Scanning requires some skill and practice to get the best image information--and to scan images correctly to fit your needs. Section 3.1 shows how to use an ordered approach to scanning that can help you control the final result and get the best possible scans. 3.2 Compensate for Dust in My Scans Cleaning dust from a scanned image is frustrating and time-consuming. Keeping dust and grime out of an image that is being scanned goes a long way toward lessening the demands of dust removal. Section 3.2 tells how to limit the amount of dust by using care, maintenance, and exercising good technique. 3.3 Select a Reflectivity for My Original Before Scanning The reflective properties of an object you scan tell you what to choose when selecting an option for reflective or transparent scanning. This is a concern only if you have a scanner with a transparency adapter. Some choices are fairly obvious, but you might want to consider options for both transparent and solid object scanning. For example, you might want to turn a solid object into a mask by scanning it as a transparent object. Section 3.3 shows what to look for by visually inspecting an object before selecting transparent or reflective scanning. 3.4 Pick a Color Setting for Scans Choosing a color setting for a scan depends on the end use of the image. Don't just select RGB because you use it most of the time. Section 3.4 shows how to make the right color mode selection. 3.5 Choose Scanning dpi It is wasteful to select the maximum dpi because it will get more information to use in the image. Selecting the proper dpi keeps file sizes in line and keeps you from having to resize images. Section 3.5 shows how to select the proper scanning dpi. 3.6 Choose a Level of Sharpness for Scanning

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 3 of 27

Using sharpness does not make a blurry image much better. In fact, some uses of sharpness might harm the images you scan more than they help. Section 3.6 shows how to select a sharpness level for scanning and why to do it. 3.7 Choose a Level of Descreening for Scanning Descreening is used to blend the dots of printed pictures--in other words it attempts to digitally remove the screening. The results vary depending on selection of the descreening level; you can achieve the best results by knowing what level of descreening to select. Descreening is only a measure to make things better; it is not an absolute solution to the problem and won't perform miracles. Section 3.7 shows how to select a descreening level based on what the image is being scanned from (newspaper, magazine, art book, or other screened printing). 3.8 Choose Black and White Points Choosing the right white and black points can be the key to getting a good scan. It is not always as simple as picking the brightest and darkest area of the image. Section 3.8 shows how to make the proper white and black point selections, and shows how to use white and black points to your advantage in getting more out of an image. 3.9 Scan a 3D Object Scanning a 3D object is a little different than scanning a flat piece of art. You have to have realistic expectations as to what can and can't be accomplished. You also need to be careful of the scanner glass and getting the best lighting for the scanned object. Section 3.9 shows how to make the best scans of 3D objects while keeping your scanner glass intact. 3.10 Scan a Translucent Object Scanning a translucent object such as a glass might produce sketchy results whether you choose to scan it as reflective or transparent. Varying scanning technique to make multiple scans of the same item helps bring out the details. Section 3.10 shows how to get a good scan from a translucent object. 3.11 Scan an Oversized Image It is possible to scan an image that is larger than your flatbed's maximum size. A few planning steps and good execution is all you need to get it done within a reasonable amount of time. Section 3.11 shows how to scan multiple parts of an image and get them together in Photoshop with the least amount of trouble. FIGURE 3.1 This is a scanning palette with ample features from LinoColor (Linotype-Hell). Custom selection and capability gives you more control over how you obtain the original image and better results in the end.

3.1 How do I...Get the best scans?

Problem

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 4 of 27

I have access to a scanner and want to get the best possible results. What are the general procedures that I should take when approaching a scan? Technique To get the best results from any scanner, you need to know what tools the scanning software offers and use those tools to your advantage. Be ordered and methodical in the preparations, and it will help you develop good technique. It is necessary to prepare for scanning in the proper way and to know what you will use the scan for. Putting all these factors together makes for the best scans. The following steps assure that you take the necessary measures to get the best image. This is an overview of the process only. You can find more detailed information on each step in the sections that follow in the chapter. Steps 1. Read the manual about the scanner's software, tools, and functions. You might do some test runs to see how the scanner's tools work and to see if the software is capable of producing results comparable to Photoshop. Sharpening, descreening, cropping, histograms, and curves or color correction might all be better left for Photoshop if the scanner shows little or no difference. If Photoshop offers better advantage, shut off the settings for those scanner features and save the adjustments for Photoshop. When reviewing a scanner's operation and software, pay special attention to the sections on calibration. A calibrated scanner yields the best results, and good software builds ICC profiles for color management. SCANNER SOFTWARE All scanner software is not the same. A few packages, such as LinoColor (Figure 3.1) offer superior color management capabilities and a full range of controls. The scanner's software can make quite a difference in the potential of the scans and performance of the hardware. 2. Before placing an image or object to be scanned, be sure that both the scanner glass and the object or image are clean and free of debris, dust, and fingerprints. Dust on the objects being scanned might require far more time and effort to correct in a scan than it would take to quickly wipe objects clean. Use manufacturer-suggested directions for cleaning and maintaining the scanner, and get dust-free soft scanner cloths for wiping objects and scanner glass. Compressed air can help blow away dust on objects and glass as well. For more information on compensating for dust before scanning, see section 3.2. 3. Place the item on the scanner in an orientation best suited to your needs. It might not always be possible to fit it the way you want. Work with the space you have. This does not limit you to scanning things that fit only in the area of the glass. (Section 3.11 covers scanning oversized objects in detail.) 4. Consider the following and set the available tools to optimize the image information when scanning. All these options might or might not be available with your scanner or software package. Reflectivity of the original: Do you want to scan the object as reflective, transparent, or file://I:\chapters\i\ix706.html 3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 5 of 27

negative? Choose the proper reflectivity. See section 3.3 for more information on selecting reflectivity. Color mode: Will the scan be used for printing in color or black and white, line art, video display, or another definable purpose? Choose the color mode appropriate to the end use. If the image will have multiple purposes, scan in a mode with broader gamut. In most instances, the scanner default will be RGB, but some excellent manufacturers (again, Linotype-Hell) rely on Lab color (as does Photoshop) for color management and conversions. See section 3.4 for more information on selecting a color mode. Resolution: Again depending on the end use, what dpi will you need? Choose the dpi appropriate to the end use. If the image is multipurpose, scan to the highest necessary dpi. For more information on choosing dpi for scanning, see section 3.5. Size: Will the image be used at 100% of the size? Do you need to size it up or down? Size the scan to the appropriate percentage for end use. If applicable, set optional features such as descreening (section 3.7), sharpness (section 3.6), gamma, and color controls (section 3.8). You can set most features after making a preview but might as well set them now while looking at the palette options. 5. Preview the image and crop it to the area to be scanned (see Figure 3.2). You don't need to scan the whole available scanning area if using only a portion of the image (see Figure 3.3). Scanning what you need will save disk space. FIGURE 3.2 An image cropped to the edges of its frame. The file size is approximately 6MB. FIGURE 3.3 In cropping the image to the area that will be used in this case, the file size is 3MB, or approximately half the file size.

6. You might want to do a prescan before doing the final scan to further evaluate the image. Some software packages have prescanning (really a preview of the final scan). If your package does not, simply drop the dpi to 72 and scan the image. This gives you a preview of the image at the size you are scanning and allows you to check the image prior to making the final scan. 7. When the prescan is complete, open the image in Photoshop and take a look at the histogram; use your eye and the Eyedropper to evaluate the relative success of the scan. Chapter 5 (sections 5.1, 5.3, and 5.6) gives you a better idea of the type of checks you should perform on the test image and can better explain how to evaluate the image. Based on your evaluations, make changes in the scanning parameters. (You might want to do color corrections, change the tonal range, fix the crops, and so on). Be sure to change the dpi back to the target range after the prescan test is done so that you can make the final scan. 8. Set white and black points (see section 3.8). 9. Scan the image and check the results in Photoshop. (Again, for help in evaluations, see sections 5.1, 5.3 and 5.6.) If the scan seems to need a lot of work, you might consider scanning a second time and correcting the problem in the scan, or you might change your approach to the scan. (See sections 3.8 and 3.10 for ideas on what to do to approach your scanning differently.) How It Works

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 6 of 27

Using an organized attack to scan an image gives you more consistent results with your scans. Get to know your scanner and software and what they can and can't do. Rely on Photoshop to help you with the rest, but get the best information you can in the original scan. Getting the best information in the original scan means less work and correction later--and less tweaking to cover up for image information that isn't there (but should have been). Comments Simple scanning variations can help you make your composite look better than an original. For example, instead of making just one scan of an image, it can be better to make two and superimpose the results, or take the best information you can from what is available in the two scans. This is very much akin to compositing photos (see sections 8.3 and 8.4), or taking several pictures of a group and using the best faces and expressions to build a good portrait. See "How do I scan a translucent object?" (section 3.10) and "How do I choose black and white points?" (section 3.8) for more information on combining two scans of the same image.

3.2 How do I...Compensate for dust in my scans?

Problem My scans consistently seem to have some sort of dust in them. What can I do to eliminate or at least lessen the problem? Technique Dust is the nemesis of anyone who does scanning. Dust always crops up whenever you scan. At times, it is accentuated by the scanner and becomes more prominent even when software adjustments are made for dust (not recommended). There is no real solution to keeping it completely out of a scan, even building a dust-free environment, wearing plastic suits, and handling items to be scanned with sterile tongs reduces only the accumulation. Regretfully, a dust-free environment does not eliminate dust on whatever it is you are scanning. Some care helps keep the dust, fingerprints, and scratches to a minimum. Cleaning the scanner and the item being scanned means less for you to correct later after the scan is done and saves a lot of unnecessary Photoshop work. Steps 1. Schedule scanner cleaning and maintenance. This means cleaning the scanner glass inside and out and perhaps cleaning the lamp and internal parts as well. Check with your scanner manual to decide on a mainte-nance schedule and stick to it. Follow the scanner instructions for disassembly (if necessary) and for any suggestions on using cleaners, solutions, and wipes. This keeps natural accumulations of dust to a minimum. 2. Wash your hands before scanning. This reduces, not eliminates, the possibility of fingerprint smudges. You might consider wearing dust-free white mesh gloves while working with a lot of scans as this keeps items free of fingerprints as well. 3. Clean the object you are scanning. Use pressurized air, soft dust-free wipes, and the sticky

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 7 of 27

edge of a sticky note (or other light adhesive tape). Depending on the object, you might just run it under water. (I know someone who used to do this with record albums and he got phenomenal results with seemingly unplayable vinyl.) Use cleaning solvents only as applicable and practical and only if you are sure they will not damage the item or cause a film or residue that will be picked up by the scanner. Be extra careful with any type of film, print, or original artwork. 4. Clean the scanner glass frequently while using it between scheduled clean-ings. Use soft, dust-free wipes. When things get really bad, you might have to resort to a glass cleaner. Just a bit on a dust-free cloth takes care of your problems. It is suggested that you don't spray the glass. Put the cleaner on the cloth, use it sparingly, and never douse your scanner with it. WARNING Do not use cleaning fluids of any kind unless suggested by the scanner manufacturer. 5. Use dust and scratch controls in scanning software only if you are very happy with the results they produce. Many such filters work simply by applying blur and sharpen to the image, which is not the best way to remove those nasty specks. 6. Rely on your tools in Photoshop to help you accomplish the rest. See Chapter 8, "Cleaning, Repairing, and Altering Images." How It Works Keeping both the scanner and object that you are scanning clean helps reduce dust, dirt, and time you need to spend later repairing and correcting images. Again, getting the better scan reduces the amount of time you have to spend correcting, rather than creating, with Photoshop. Comments Sometimes, a simple wipe at the right time saves hours of correction. Suppose, for example, that in scanning a pile of 100 or so 8x10 photos, you place the first one on the scanner with the wrong orientation. In getting it off the scanner, you barely touch the glass, but your finger leaves a mark. You choose not to notice it because a wipe isn't handy. When you are done with the pile, you spend an average of an extra minute on every photo cleaning off the same smudge. That is an hour and a half of wasted time when you could have spent 30 seconds with a wipe. However, there is a point of diminishing returns, where time spent cleaning the object and scanner will exceed the time it takes to correct the problem in Photoshop. Obviously, that is the time to stop cleaning and start scanning. Work to get rid of the obvious dirt and dust before scanning and leave the rest for corrections. Some very good Photoshop tools might make quick work of the problem (such as proper use of the Dust & Scratches filter; see section 8.2).

3.3 How do I...Select a reflectivity for my original before scanning?

Problem

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 8 of 27

There are several options for the type of image being scanned, including transparent, negative, and reflective. Which type do I pick? Technique Unless you are going for some type of special effect, the choice of scan type should be relatively simple based on observation of what you are scanning. The names might vary depending on the software packages used for scanning, but either you will be dealing with a solid object or a transparent one. If transparent, you might have to choose negative (such as traditional negatives for making photographic prints) or transparent positives (slides, chromes, transparencies, and so on). The correct choice will get results. Some objects will prove very difficult to scan as either reflective or transparent--usually a problem with clear objects. Scanning clear objects is covered in section 3.10 later in this chapter. Steps 1. Look at the object you want to scan. If the object is solid and you are scanning the surface, choose reflective. Your choice here is to have the scan appear as light reflects from its surface. This works for paper prints, photographs, text on paper, fabrics, and any physical object that fits on the scanning plane. Other examples might be hands, faces, utensils, and so on. In other words, the scanner can be used as a rather limited digital camera for three-dimensional objects as well as scans of flat art. WARNING Be careful not to scratch the scanner glass, especially when scanning objects. See section 3.9 later in this chapter. If the object is a film strip or translucent, and the color is a positive representation of the image you want to use, choose transparent. This works for slides, chromes, movie film, and other positive emulsion film (where color appears normal rather than negative). You might also consider it for translucent objects or creating silhouettes of solid objects (see the example in Figure 3.4). FIGURE 3.4 A silhouette of a solid object, such as this slotted spoon, is easy to create with a transparency adapter. Simply scan the object as a transparency (positive or negative). If the object is a film strip or translucent and the color is a negative representation of the image you want to use, choose transparent negative. This setting might also be used to create some unusual effects in scanning translucent objects. 2. Be sure all other parameters for the scanner are set properly and complete the scan. How It Works Choosing the proper object reflectivity tells the scanner how to scan the object. Choosing the proper setting is important for getting the right effect but also might change the settings for other options. For example, changing the reflective setting to transparent will probably disable the descreening, as

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 9 of 27

descreening is only useful in capturing screened prints (which are reflective). Note the differences in the scanning choices and make the best use of them. Comments Though this might seem relatively obvious, the results of choosing the wrong object type can prove frustrating. Consider the possibilities of using other technologies for images that are just too challenging for your equipment. For example, you can get very good inexpensive scans from PhotoCDs for all your 35mm needs, and you might use those even if you have a reasonably good scanner with a transparency adapter. To a point, PhotoCD scans rival drum scans and are a small fraction of the cost. Getting a scanner with a transparency adapter and paying a lot for it is probably a mistake unless you plan to be scanning medium format transparencies, larger chromes, or medium format slide film (6cmx6cm and above) or plan to use the adapter for other creative purposes--or just for fun. Regardless of the maker or resolution, I have yet to see one perform to the level that would produce a good sharp scan from any 35mm film image with the hope of good reproductive quality. For 35mm negatives and transparencies, slide scanners are far better than an adapter. Again, PhotoCD or other similar scans might be a less expensive option. There are many opportunities to use a scanner creatively. The choice of the type of object you scan might not be as simple in some cases as it seems. Some objects call for creative techniques in scanning. For example, I used additional lights for scanning dark 3D objects. Figure 3.5 shows a scanner set up with a spotlight and aluminum foil reflector. A camera lens was scanned by placing it in the aluminum cave, which helped surround it with light to pick up more detail and depth. Be careful of creating color casts using lights not balanced to the temperature of the scanner bulb. Correction becomes difficult when using two different color temperature light sources. FIGURE 3.5 A setup with the scanner to add supplementary light to 3D object scans. SCANNING WITH ADDITIONAL LIGHTING In using additional light with a scanner, be aware of the color temperature of the added light. If using incandescent light (as pictured in the example for 3.5), the areas where the light strikes from the spotlight will cast red. This is because the scanner will be calibrated to the scanner bulb that is rated at a bluer color temperature (about 5500°K compared to 2700°K for incandescent), unless you make changes in the calibration-which is not suggested. Instead, match the rated color temperature for the scanner bulb with added lighting to achieve spotlighting without color casts. Check the scanner manual to be sure of the color rating for the bulb. It is easy to make a mask, frame, or silhouette out of almost anything that will fit on the scanner glass. Figures 3.6 and 3.7 show the result of using everyday items to make unusual masks for images. The mask in Figure 3.6 is made by making a transparent scan of a solid object, and then the resulting mask is applied to frame the photo in Figure 3.7. Crystal and other translucent objects might produce interesting results when scanned as transparencies as well. FIGURE 3.6 A pile of cotton swabs scanned using the transparent setting. FIGURE 3.7 Taking the scan from Figure 3.6, it is easy to make a mask and apply it to frame an file://I:\chapters\i\ix706.html 3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 10 of 27

image. If you didn't know the frame was made of cotton swabs, it might look more like a spatter or droplet effect. (Violin by PhotoDisc) TIP When operating a scanner, remove jewelry (especially anything with diamonds, which can inadvertently scratch the glass). When glass is scratched, you are doomed to seeing the scratch in every scan you make.

3.4 How Do I...Pick a color setting for scans?

Problem There are several settings on the scanner software for different color modes. How do I choose between the settings? Technique The technique here is as simple as selecting a button, but choosing the right one is a matter of looking objectively at the thing you want to scan and knowing the purpose you will use it for. Sometimes, the RGB 36-bit color option is not the right way to go just because it will get more information and deliver better potential color depth. If you are looking for portable images and a quick scan, and you will be reducing the size of the original considerably, your choice in scanning technique might need to be based on utility and not optimum quality. Steps 1. Considering the use of the image, select from the color options list in the scanning software or plug-in. The following options might be available: Line art (or bitmap), grayscale, halftone, RGB, RGB (X)-bit, CMYK, CMYK (X)-bit. FIGURE 3.8 The Grayscale conversion of Figure 03CDfig02 from LAB mode. Compare this with 03Cdfig02 from the CD-ROM. NOTE "X" can be various numbers depending on the claims of the scanner maker or software manufacturer. Eight bits per channel actually translates to 24-bit color in RGB, or 8 bits for each channel of the three colors (8x3=24). Twelve bits per channel scanning in CMYK is 48-bit color (12x4=48). Impressive depth, but many Photoshop features are disabled using images with more than 8-bits per channel. (Photoshop 5 has added more support for images with up to 16 bits per channel; see Adobe's release notes on Photoshop 5 for a list of features.) You will generally end up translating images into 8 bits per channel to use full Photoshop features. Even if you could work on the images with better depth, the increased size of the files (approximately 50% larger for 12bit/channel images over 8-bit/channel and approximately double for 16 bit/channel images) might prove prohibitive to your equipment if the image is large. BITS PER CHANNEL VERSUS BITS PER PIXEL

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 11 of 27

The number of bits per pixel is the total number of bits for all the channels that make up the pixel information. For example, RGB has three color channels (red, green, and blue), each of which has a specified bit depth (usually 8 bits). To get the total bits per pixel, multiply the bit depth by the number of color channels (16x3=48 bits per pixel). Grayscale has one channel, RGB has three, CMYK has four, and spot colors are additional. To check the bit depth of the channels in an image, look at the Mode menu (Image, Mode). TABLE 3.1 Choices and Suggestions for Choosing a Scanning Mode SUGGESTIONS This works for most color images and ink type drawings (which are not mixed media), text only, single color logos, and so on. The advantage of using bitmap over grayscale is you can use higher resolution and retain a smaller file as there is less bit depth--approximately 1/8 of the information to deal with in a similar size Grayscale scan. You can use twice the resolution, have a smaller file size, and cut a nice sharp edgeline. Choose grayscale if you This works for most color images and might have some creative use in are scanning to use an developing textures (for example, scanning fabrics). The advantage here image in B&W grayscale. over RGB is in saving disk size. Some color images are better scanned in RGB even if the intent is to use them in grayscale as the tonality of the colors is too similar to provide proper separation (see Figures 3.8 and 3.9 and 03Cdfig01 on the CD). Choose RGB if you are This will be the choice probably 90% of the time. Scanning to RGB even scanning an image for if the image is going to be grayscale might offer opportunities to fix the multipurpose color and image before converting to grayscale. general accuracy. Choose RGB in deeper bit Without the capability to use the extra information, this will waste file mode if you will be using size. Most Photoshop features will not work with 16-bit+ images. the image in applications where extra bit depth matters. Choose CMYK if This gives you a good color approximation of RGB color within the scanning images strictly range of CMYK and accomplishes it with less fiddling than will be for print reproduction. required if the scan is RGB. Results will vary depending on the scanning package, calibration of the scanner, and generation of correct ICC files. See your scanner and scanner software manuals. Choose CMYK in deeper Again, many Photoshop features will not work in 16-bit mode. bit mode for specific application. FIGURE 3.9 The Grayscale conversion of 03CDfig02 from RGB. The better grayscale conversion of 03CDfig02 is achieved through RGB in this case, but the results can be reversed with different color combinations. COLORS SCANNED TO GRAYSCALE CHOICE Choose line art or bitmap if the item you are scanning is pretty much hard black and white.

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 12 of 27

The Grayscale conversion comparison in Figures 3.8 and 3.9 shows how color images can appear to flatten if scanned in Grayscale without additional color correction. If the tonality of the colors is exactly the same, conversions will flatten in LAB. A similar thing can happen in scanning to RGB (it just isn't as easy to demonstrate). To test it out, try to do conversions from different modes with 03cdfig02 from the CD-ROM. 2. Be sure the other parameters for completing the scan are set properly and complete the scan. How It Works Selecting the proper mode for your purposes keeps file size within a proper range and helps target your output or usage. Getting sufficient information is critical to producing good results in the shortest amount of time and with the least trouble. Poor choices in color selection lead to additional color work and wasted time and can cause unnecessary file size problems. Comments If you can target your scanning, you can scan more efficiently and can make the proper adjustments to get the best results. For example, choosing CMYK to make a scan that will be used in grayscale is wrong for just about any purpose; you get too much information and a larger than necessary file size, and correction for tonality and color would be easier in RGB. Stick to your purpose and it will yield the best results. For general scanning, use RGB. Calibration of the scanner is essential for getting good scans and good CMYK conversions. Some might insist that RGB scanning is the only way to go--and to use maximum bit depth--but files not produced as close as possible to their final intent still need to be converted. In other words, as high bit depth is not supported by all types of output (or all programs that you might be using the images in), scanning with the highest bit depth might just mean you will have to convert the image anyway; though it is technically a better scan, it might not be best suited for your end use. Also, any additional conversions lead to a generated loss or alteration of information in the scan. Some alterations are potentially positive, but most are not. Let the software work for you and do your best to help it do its job by properly targeting your output.

3.5 How do I...Choose scanning dpi?

Problem With limitations set only by the range of the scanner, how do I know what dpi to select? Technique As with choosing an image dpi, you want only as much information as you will use for the scan. Knowing an end size is the optimum way to scan, but again, if an item is going to be used for several purposes, you might want to scan with maximum potential for the resolution in mind. For example, say you are scanning an image for a Web site that requires only 72dpi. However, consider a situation where you know you are also going to use the image in a color magazine advertisement at a similar size later on. The magazine (printed at 175 line-screen) might require between 271 and 350dpi, which is approximately four to five times as much image information (see sections 1.3 and 1.5 for more

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 13 of 27

information on line-screens). You can scan and store the image separately for each, but it might be easier (and provide more consistent color) to scan for the two purposes at one time. It might be prudent to choose to scan at the greater of the two resolutions. The other choice is to make two targeted scans--one for the Web site and one for the magazine. You need to determine which set of options and trade-offs best serves your purpose. TIP When scanning, remain within the limits of the scanner's optical range and don't use interpolation if it isn't absolutely necessary. Interpolation is simply electronic resizing, which Photoshop can do for you if you choose that route. If you need a larger scan than can be provided by your scanner, consider having the scan done for you. This might require using a service bureau or other imaging professional. Steps 1. Determine what you want to use the image for. 2. Determine the dpi needs for the output/use of the image as outlined in step 1. The scanning resolution (ppi/dpi) should equal the dpi necessary for the end use (see sections 1.3 to 1.5 for more information about selecting dpi). 3. Set the resolution in the scanning parameters. Be sure all other scanning parameters are set and complete the scan. How It Works Optimally the scanning resolution equals the output without resizing. A scan can be larger and resized down, but there is no reason to do that if not necessary. It is easiest and most efficient to keep images at the size needed. Comments If you are unsure of the final size and dpi that the image will use, this creates a problem in selecting scanning parameters. If the scan is going to be a utility image--that is, one you will use for various purposes over a period of time--you might consider scanning and storing the image at several sizes to have it ready whenever needed. Kodak PhotoCD scans, which offer multiple scan resolutions, are well suited to this purpose. If you plan to keep a library of source images that you will reuse, a more permanent type of removable storage might be in order so you are not left with huge libraries of images clogging up your hard drives. A CD-R drive offers reliable long-term storage and compatibility--both with different operating systems and hardware, large disk capacity, and compact size, and they are inexpensive--for both media and drives. TIP Though technologies are changing all the time and DVD recordable drives are probably not far off to use in data storage, CD-R drives currently offer the best combination of capacity, compatibility, and cost. The disks, if formatted correctly, work well even across platform in standard CD-ROM drives. The media is very inexpensive compared to any other storage source in cost per megabyte.

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 14 of 27

DVD DRIVES DVD (Digital Video Disk, sometimes referred to as Digital Versatile Disk) drives are much like the current standard CD-ROM drives but with two layers of digital information on a disk side. This leads to a big difference in storage capacity: DVD has 7 to 13 times the storage space on a single-sided disk than a standard CD-ROM has.

3.6 How do I...Choose a level of sharpness for scanning?

Problem I am making a scan and want it to be sharp. How do I use the sharpness control in my scanning program? Technique The intensity of the sharpening tool on your scanner software might not function the same way as Photoshop's tools or the same as the sharpening tools in other scanning programs. It might be best to leave the sharpening for later and just do it in Photoshop, rather than use tools that are not as good as those in Photoshop. However, if the tools are available and they perform well, you might want to use them. Keep the intensity of the sharpening low so that sharpening enhances rather than damages your images. You can always go back to the image later and sharpen it more with Photoshop if you want to increase the effects. Become familiar with the way sharpness works with your scanner and software by running several tests on the same image. Over-sharpening causes halation (a noticeable glow in areas of high contrast) and can accentuate film grain and make flaws stand out more prominently. Figures 3.10 and 3.11 demonstrate the effects of over-sharpening. Figure 3.10 is sharpened using a low radius and medium (50%) sharpening intensity. Figure 3.11 shows sharpening with too high a pixel radius and too much intensity. Steps 1. Place the image on the scanner and open your scanning tool. 2. Select a low level or no sharpening from the list. 3. Be sure all other options are set correctly and make your scan. How It Works Sharpness accentuates the edges of tonal change by affecting local contrast. In other words, when sharpening, a dark tone that aligns with a light tone gets darker, and the light tone gets brighter. As shown in Figure 3.11, this can cause halation and other image damage so should be used with care or not at all. If the tools in Photoshop offer more control than the scanner tools (and the Unsharp Mask is quite a versatile tool), sharpen the image in Photoshop. Waiting to use Photoshop for sharpening has the additional advantage of leaving sharpening until after an image has been relieved of damage and dust. Sharpening first accentuates the damage and dust in an image.

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 15 of 27

FIGURE 3.10 Sharpening this image builds slightly better contrast and separation, which can also help offset the effects of dot gain in printing. Original image by PhotoDisc. FIGURE 3.11 Over-sharpening makes the image unnaturally contrasty and damages image information. Note the halation around the subject's hair--this is not desirable with this intensity. Comments Generally rely on Photoshop for sharpness corrections because scanner control of sharpness might be vague and Photoshop's tools are fairly precise. Use this rule of thumb especially when a software package offers only Low, Medium, and High as selections for levels of sharpness. Even if they offer percentages, it is hard to say what scale they are using to measure those percentages. The scanner's advantage is in providing raw data. Let it do what it does best.

3.7 How do I...Choose a level of descreening for scanning?

Problem I am scanning an item from a printed piece and want to use descreening to eliminate the effects of printed dots. How do I know what line-screen to choose in setting descreening? Technique Considering the paper an image was printed on and knowing the source of the image will give you a good idea of what line-screen was used in printing. Simply selecting a descreening level based on the paper type can give you a good idea of the line-screen and will generally deliver the results you need. Paper might appear to be absorptive, like newsprint, or might hold dot gain to a minimum, like a heavy-weight coated cover stock. When using different paper stocks, printers use a certain range of line-screen to get the best results based on known absorption and dot gain. Understanding the choices a printer makes can help you know the line-screen he printed at and infer a descreening level. NOTE Descreening is not an absolute solution to smoothing out the effects of printing an image and making it into a continuous tone. It is merely an aid to bettering the potential. See Figures 3.12 through 3.15 for examples of the effect of descreening an image. CAUTION Do not use descreening on images that were not printed with line-screens unless you want to experiment with effects. The resulting image will be blurry, soft, and generally unpleasant. Steps 1. Look at the image source and evaluate it depending on the following criteria. TABLE 3.2 Approximate Line-Screens Based on Paper Type and Image Source

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 16 of 27

PAPER TYPE AND IMAGE SOURCE Newsprint A book or publication on uncoated paper stock (or something that looks rather rough and absorptive) A book or publication on coated paper A book or publication in color on smooth stock An art book, art publication, or collection

APPROXIMATE LINE-SCREENS 75-120 lines 120-133 lines 150 lines (a common line-screen for print half-tones) 175 lines (includes color brochures, magazines, book jackets) Up to 200 lines (includes Duotone images and museum- quality books)

2. Be sure the other parameters for the scan are set and complete the scan. How It Works Descreening diffuses halftone dots and smoothes the tonality of an image using a combination of blur and sharpen effects. The results are not often completely desirable, but the alternative (no descreening) is often worse. You can accomplish similar effects with Photoshop tools but probably not without significant trial and error. Using descreening will probably save some time and heartache in these cases. If the lack of quality is not acceptable for your purposes and the screened image is the only source, use another image. Comments Unless the original image is unavailable or you are seeking to create an effect, scanning a printed image should be a last resort--even using descreening to improve the results. As with any type of image integrity, each step removed from the original loses measurable quality. Scanning from a printed image will prove somewhat extreme as far as steps removed from an original goes, as the image that you have seen in print has already been scanned, prepped for printing, and printed. If at all possible, getting closer to the source will yield better results. FIGURE 3.12 A close-up of a portion of an image scanned from a screened print with no descreening. FIGURE 3.13 The same screen-printed image as in Figure 3.13 with 133 line-screen descreening. FIGURE 3.14 Figure 3.12 printed at the intended size. FIGURE 3.15 Figure 3.13 printed at the intended size.

3.8 How do I...Choose black and white points?

Problem I am not sure how to choose black and white points in the image I am scanning or why I should do it. Is it necessary? If so, what is the best method for choosing?

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 17 of 27

Technique Scanning to particular black and white points can be either fairly arbitrary or rather precise depending on the capabilities of the scanning software and the image, or what you are trying to accomplish. A good scan will get as much information from an image as possible without going to the point of compromising tonality, that is, high-key or low-key images need to be scanned with shorter tonal range to catch the detail you will need in order to reproduce them faithfully. If it is possible, adjust for the type of image being scanned before making a final scan. TIP Even if your scanner has a preview, looking at the preview will probably not tell you all you need to know about the image in order to scan it well. It might be best to prescan an image if it looks like a challenging one. To prescan the image, scan at 72dpi. Using Photoshop's tools (notably the histograms), determine how it is best to scan an image based directly on the information from the rough scan (and information on evaluating images from sections 5.1, 5.3 and 5.6). Be sure the scan concentrates on the main tonal range of the scan without forsaking detail. The real key here is that your scanner software has the capability to enable you to adjust for the scan. Without that, considering black and white points is moot. Steps 1. Place the image on the scanner. Preview the image. You might also consider prescanning the image at a lower resolution. 2. Use the scanner's tools to adjust black and white points for the image based on the scanner's sampling tools or adjust as per the prescan. To select a white point, select from the darkest part of the image that you want to be considered as absolute white. To select a black point, select the lightest portion of the shadow areas that you want to be considered as absolute black. Your selections should, optimally, not remove important detail. 3. Be sure other scanning parameters are correct and complete the final scan. How It Works Choosing the black and white points for the scan limits the range where the scanner looks for information. Initially this might sound like a bad thing--why limit the scanner? But the limitation on the range enables the scanner to concentrate on the detail and bit depth over the rest of the image. Tailoring the scanning range to the tonality of the image specifically allows the scanner to make the absolute best scan considering the range of information in the image being scanned. Comments It is not necessarily bad to scan images without consideration of black and white points, but you will get better results if you do consider them.

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 18 of 27

TIP Unless your scanner software is highly touted and has excellent cast correction attributes, just make the scan and do the color corrections in Photoshop. Chances are that Photoshop's tools are better. As stated previously, it might be advisable to make more than one scan of an image using different settings for black and white points, especially if the image is a difficult one (see Figure 3.16). The technique of multiple scanning can be demonstrated well by using a high contrast photo where the subject is backlit. In this case, make one scan with the intent of getting good general image information (scanning what the image was exposed for) and another trying to bring detail out of the shadows. In the first scan, set the white and black point as you would normally. In the second, set the white and black point to the area of the image that has fallen into shadow due to the backlighting. This gives you the general tonality from the image in the first scan and some better detail from the shadows to blend in and improve what would be available by incorporating the second scan. The composite of the two scans can be accomplished fairly easily using layers, layer blending, masks, and other advanced techniques covered later in the book. A key to success is making the second scan without touching the image in between, that is, make the second scan just by changing the scanner parameters for black and white points without changing the cropping or placement of the image on the scanner. In doing this, the images will be exactly the same size and will be perfectly aligned when brought together as separate layers in a document. Depending on the image, you might find yourself doing several scans of a single image to fully extract the potential information available there. Figure 3.16 shows three distinct tonal ranges. Specular highlights are apparent from the sunlight coming in and striking the sheets. The sheets in room light show everywhere from white to almost black in the deepest shadows. The shoes are certainly in shadow, yet beg to project a bit more detail. The results of the image are due to the limitation of film and equipment to capture the image. In other words, being in that situation, the bright and dark areas would not completely lose detail as they do here if looked at with the human eye. Scanning for the highlights, overall tonality, and shadows separately will help produce the best results. Figures 3.16 through 3.19 show how to get better information over an isolated tonal range. For sake of brevity in the example, the discussion is limited to the midtones and shadows. Scanning in this way helps you get more out of the original and helps you bring more to Photoshop in order to get the image results you want. FIGURE 3.16 A difficult image exposed for the midtones, yet having extreme highlights and shadows. FIGURE 3.17 An original portion of the scan from Figure 3.16 shows a lack of detail in the shadows. FIGURE 3.18 Scanning a second time to get the detail in the shadows gives you the extra you need to improve the composite. The scan looks bad overall, but you are concerned only with the small part of the image that shows shadow detail. FIGURE 3.19 Figure parts from 3.18 and 3.16 are composited and blended using layer blend, selective erasure, and tonal control via curves. This is a better representation of detail that might

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 19 of 27

otherwise be lost without multiple scans.

3.9 How do I...Scan a 3D object?

Problem I want to scan a three-dimensional object, but I want to keep it from scratching the glass and don't quite know what to do with the scanner lid. How do I scan an object without compromising my equipment yet still get decent results? CAUTION Hold on to that transparency adapter lid when scanning 3D objects or you might end up in a disaster of broken glass. Technique Generally, pull up the top of a flatbed scanner and place the 3D object where you want; then scan. However, to keep the glass from being damaged or scratched and to keep it clean, it might be a good idea to separate the object from the glass. This actually provides a two-fold benefit depending on the depth of the object: first, you protect your glass, and second, you get better, more even lighting of the object when it is not overly close to the scanner light. A little distance lets the light disperse more evenly, rather than harshly lighting the areas of the object that are closest to the glass. Transparency plastic (used with overhead projectors) works well to protect the glass, but something that creates a bit more distance between the object and the light might actually work better. Obviously, you need to use something that is completely transparent and large enough to fit the item you intend to scan. The one drawback to placing something between the object and the glass is that when you have more surfaces, more places can accumulate dust and dirt. Steps 1. Prepare for scanning by wiping the scanner glass. 2. Place the protective barrier on the scanner glass. Be sure, of course, that it is clear of dust debris and scratches. 3. Place the object on the protective barrier. Orient it to get the view you want from the scanner's perspective. It looks up at everything (from the base) when you are scanning reflective, and down (from the transparency adapter) when scanning transparent. 4. Set the scanner features. 5. Complete the scan. How It Works Placing a barrier between the object and the glass protects the glass and might actually help with scanning. Clear Plexiglas glass will work, as will any number of other clear items. Try to keep the

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 20 of 27

surfaces as dust-free as possible. A personal favorite for glass protection is using a CD jewel case cover (obviously with the insert removed). This will raise the object about 1/4 inch off the glass, is transparent enough to give reasonable results, and is usually readily available around computers. The drawback is that it only works with small objects. Comments As shown earlier, it is possible to scan an object to create a frame or mask, but it is also often desirable to scan the object itself for incorporation into other art or to use as a graphic. In Figure 3.20, A small gargoyle statuette was placed on a new CD-ROM case over the scanner glass and balanced on its ear. After scanning, a little extra clean-up was necessary because two additional surfaces were placed between the object and scanner eye. The more surfaces, the more dust and scratches. However, the results are certainly satisfactory. There was another attempt earlier to complete this scan without anything between the object and scanner glass, but it just didn't work-light was too intense on the parts of the object closest to the glass (the ear and nose) and the image detail washed out irreparably at those points. Tricks to reduce dust and scratches helped to make this painless. (See section 8.1 and 8.2 for information on removing dust and dirt from scans.) FIGURE 3.20 A small gargoyle statuette scanned on a new CD-ROM case cover. You might keep a few fat computer books (such as the one you are reading) on hand to help prop up the scanner cover when scanning objects. Especially with a transparency adapter, propping up the cover will keep pressure off the scanner glass. Using a black-and-white background (depending on the color of your objects) will help separate your scanned objects from the rest of the image area and can help you make selections, masks, and clippings more easily. Use a black background with light objects and a white background with dark objects. Lay the backgrounds over the object after it is in place. A background can be anything from paper (plain old typing paper) to fabrics (black velvet can provide good separation, though it is a dust magnet) and can actually help to hold objects steadily in place. Aluminum foil might be useful in helping to provide more even lighting around an object.

3.10 How do I...Scan a translucent object?

Problem I want to scan something made of glass. I tried reflective scanning, but the image is far too faint. Is there anything I can do to get results? Technique The results of scanning transparent objects can be interesting if you use the right objects and the right techniques. The technique described here might not work for all transparent objects, but it will add depth to almost anything that you might have difficulty scanning conventionally because of its transparent nature. Transparent objects might be glass, items made of clear plastic, and so on-anything that you can see through.

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 21 of 27

The idea is to take advantage of an object's translucence, or its capability to deflect light. Superimposing a transparent scan with the reflective information can yield results that would be impossible with other straightforward reflective or transparent scans. The information from the two scans put together can reveal a better visual representation of the object. Figures 3.21 through 3.31 use the steps to complete a rather difficult scan of a plain wine glass. The reflective scan will normally show some reflective highlights that the transparent scan would not, and the transparent scan will show some interesting light deflection that would not occur in a reflective scan. Making both scans and putting them together gets the best overall detail and result. See 03Cdfig01 on the CD-ROM for a closer look at the example. NOTE Do not move the object or change the cropping when using this technique to composite a scan. If you do, the parts will not match. CAUTION Though a scanner might allow some limited play in detaching the transparency adapter (Linotype-Hell scanners offer some 3-4 inches of play in the detached state with the supplied cord), some adapters function only when secure. Be very careful in operating the transparency adapter detached from the base and be sure it is secure as you situate it; scanner vibration might cause the detached adapter to move or settle during operation and the results can be disastrous. Steps 1. Be sure the item you are scanning is clean. Handle it with gloves if necessary to keep fingerprints off. 2. Scan the item with normal reflective settings. Depending on the object and your scanner, you might want to provide a background. In Figure 3.21, the scan was made using a flat white background. Scanning with different backgrounds produces very different results, especially with clear objects, which see through the object to the background. FIGURE 3.21 Reflective scanning leaves little to see in the wine glass but some highlights. This scan would probably not be acceptable for most purposes and would require a lot of manipulation if it were intended as the end result. 3. Save the scan. 4. Scan the item again, this time using transparent scanning (see Figure 3.22). FIGURE 3.22 A scan of the wine glass done using the transparent scan setting. The glass bends and reflects light as it passes through, and produces this slightly soft but rather dramatic scan of the glass. 5. Save the scan.

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 22 of 27

6. Copy the first scan into the second scan on a new layer. This can be done in a variety of ways. The most straightforward is to select all, copy, change documents, and paste. 7. Reduce the opacity of the upper layer. The upper layer will be the scan you just copied and pasted into the document in step 6. You should be able to see both the glasses (see Figure 3.23). They might be slightly different in size and orientation, depending on how they were situated on the glass and the orientation of the transparency adapter. Fit the items together the best you can for now (see Figure 3.24). FIGURE 3.23 The opacity of the pasted layer is reduced to 50% so you can see how the glasses compare. FIGURE 3.24 Here is the result of the basic superimposition. The steps that follow will show how to pull out the detail and shed the unwanted image information. TIP You can adjust opacity in several ways. An easy one is to highlight the layer for the reflective scan and either enter the number in the box for opacity or click on the arrow at the right of the box and use the pop-up slider. 8. If the reflective scan does not fit well with the transparent scan, use sizing, transform (Edit, Transform), and other tools to make it fit as well as you can (see Figure 3.25). The transparent scan will carry the bulk of the importance in this example. Overlapping or extraneous information from the reflective scan will be cut off. Actually, it might be better if the reflective slightly overlaps or goes outside the bounds of the transparent scan so that information will go to the absolute edge of the cropped image. 9. Make a selection of the space around the transparent scan. See Chapter 9, "Selection, Isolation, and Masking Image Elements," for more information as to how to accomplish this. FIGURE 3.25 The Transform tools help you manipulate the reflective scan well enough to fit over the transparent scan. FIGURE 3.26 The selection of the glass was quickly accomplished using the Magic Wand tool with a tolerance of 1 and antialiasing. 10. If you want the selection tighter, consider doing that by expanding the current selection. Note that the selection in this example is already very tight to the top of the glass. Expanding the selection forces the top edge of the selection inside the transparent scan. In the case of this example, save the selection first so you can merge that top portion of the selection with the selection you create by expanding the current one. If necessary, make other adjustments to the selection. 11. Apply a soft feather or blur to the selection (3-5 pixel radius). This keeps the selection from being too hard-edged. FIGURE 3.27 When the selection is complete, it should be solid and fit rather tightly over the

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 23 of 27

transparent object. 12. With the selection in place, choose the reflective layer and then select Clear from the Edit menu. This clears all the information outside the area of the transparent scan. 13. Open the Layer Options dialog box by double-clicking on the reflective layer. While viewing as much of the image as possible onscreen, work with the sliders to produce the most pleasing blend of information. Then consult Figure 3.31 for the settings I chose to complete the image. THE BLEND IF SLIDERS Blend If sliders are a powerful image editing tool--seldom discussed, yet quite useful. Watch out for some extreme reactions and pixelation, which you can usually cure by splitting the sliders. To split the sliders in the Blend If dialog box, hold down the Option key (Alt key on Windows) and click on the right or left portion of the slider and move it. Detail outside the range of the sliders will not blend. Splitting the sliders makes the area between the split points blend as a gradient. FIGURE 3.28 With the area outside of the transparent scan cleared from the reflective scan, the image starts to take better shape. All that remains is to blend the layers for best results. FIGURE 3.29 This is a rather liberal blend and should produce a smooth mix of information in the two layers. How It Works Using the best information from the transparent and reflective scans resolves this image problem quite nicely. You might like more or less detail from the reflective scan, but you can control that any way you like using the Blend If option and layer opacities. This technique might need to be adapted to work in scanning other clear objects. FIGURE 3.30 The final result of the composited wine glass scans. FIGURE 3.31 A side-by-side comparison of the original reflective information, the original transparent information, and the composite of the two, viewed left to right. Comments This is a variation on the use of scanning twice as we saw in "How do I choose black and white points?" The theory is essentially the same: Get what you can out of the information available by doing more than one scan of a single item. Creative use of the scanner's capabilities can yield results that are quite unexpected--and sometimes very pleasing. A little planning might help you get results under control, whereas creative scanning techniques offer many image possibilities you might have otherwise dismissed as impossible. NOTE Scanning an image in the fashion described in this example will distort it, much like a wide-angle lens. It is best to keep the object centered on the glass if you are looking for

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 24 of 27

the most faithful reproduction. Sliding it to the side of the scanning area (picking it up first, of course, to avoid scratching the glass) increases distortion.

3.11 How do I...Scan an oversized image?

Problem I have an image to scan that is larger than the available area on my flatbed. Is there a way I can scan it and get good results? Technique Another two-scan technique can help you get one complete image from an item that is too large to be scanned on your scanner. You need to divide the scan into two parts (or more if the image or object is larger) and scan them separately; then put the image parts together. The scan for the wine glass in the previous example was actually constructed in two parts, as the transparency adapter on the machine I used was not detached from the base and I had no way to get the object into the scanner to scan in one pass without detaching it. TIP Use the ridges at the edge of the scanner glass to help keep multiple scans parallel. On flat items, secure sticky notes or tags on the back at a point where the scans will overlap and make crosshair targets to help line up the scans later. These junctures should also be scanned on the same portion of the scanner to help keep distortion to a minimum. To do this properly, it might be necessary to flip the image (see Figure 3.34). Care in keeping the orientation of the image and object the same keeps you from having to make more difficult adjustments later. Steps 1. Determine how many scans you want to make of an object to get it into your computer (see Figure 3.32). For example, if your scanning area is 8x14 and your image is 12x21, you need to make at least three scans. If the image is 14x16, don't try to get it done in two passes. Plan to leave about a 1-inch overlap so that putting the image pieces together will be easier. Space gives you the ability to better align and doctor the blend of the two parts. FIGURE 3.32 An attempt to scan the wine glass at the outermost reaches allowed by the scanner with the transparency adapter attached. Information broke up on either side of the scanning area, obscuring the image, so the scan was made in two parts instead. 2. Determine which edge you will align to in order to keep the image straight. Scans from a large sheet of paper (where you can be reasonably sure the edge of the paper is straight) will be fairly easy to keep in line. Something without a hard edge (fabric, statue, and so on) requires more work after the scanning is complete to align and blend properly (see Figure 3.33). NOTE Even if you are diligent about keeping the items in line for multiple scans, some adjustments in Photoshop might be required. However, the less adjusting you need to do, file://I:\chapters\i\ix706.html 3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 25 of 27

the better your results will be. FIGURE 3.33 Keep the edge of what you are scanning parallel to the side of the scanner glass. This will help you align the parts later. 3. Disable all automatic color and tonal functions for scanning. (Consult your manual for the features and how to shut them off.) You will want the scanner to scan exactly the same way in order to get information that is the best match. If your scanner has an option to choose the settings from the previous scan, use it in all successive scans for the parts of the image you are trying to rebuild after you have made the first portion of the scan. 4. If the object is flat, be sure you have placed targets appropriately with tags or sticky notes to help with the alignment. TIP When placing targets, I like to use a double crosshair: one long line that runs parallel to the image and two short perpendicular strokes (see Figure 3.34). FIGURE 3.34 Scan flat images face down with sticky note targets attached to the back. The figure on the right shows how you might need to flip the image being scanned to keep the overlap in the same area of the scanner to reduce potential distortion. 5. Make the scans in sections, being sure to overlap by about an inch. Number them and save as appropriate. If there are several pieces, you might want to sketch a map as to how the pieces should go together. 6. Take the first scan in the series and enlarge the canvas size to the appropriate dimensions for the size of the final image (see Figure 3.35). 7. Cut and paste all the pieces into the large canvas (see Figure 3.36). If the image was rotated to scan, rotate the canvas back (Image, Rotate Canvas, 180°) before cutting and pasting. FIGURE3.35 As the glass was at least 6 inches tall and there might be some distortion in the scan, the canvas was enlarged to a little over 8 inches, leaving plenty of room to maneuver in. You will want your canvas size to be a bit larger than you expect the final image to be so you can work better with movements, transformation, and rotation of elements to make things fit together. FIGURE 3.36 The other part of the image is roughly pasted into the bigger canvas. Now it is time to start trying to fit the pieces together. 8. You might want to save this file with another name so you will not save the image over your first scan. 9. Take the pieces one at a time and fit them to the original. To do this accurately, magnify the areas you are connecting to several hundred times their size. Change the opacity of the upper layer (the one you are working to maneuver into place) to 50% to see through it to the layer below. Use rotation and movement tools to get the images to match as well as you can. This is where

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 26 of 27

the targets can come in handy in alignment. Make alignments in an orderly fashion (for example, work right to left and top to bottom if there are multiple pieces) and be sure you are completely done placing one element before moving on to the next. This will save problems later with readjusting, which might cause a domino effect of changes: Changing image part 3 after placing part 10 might cause you to have to rearrange the whole image. If necessary, use transformation tools to alter the image parts for a better fit. Transformation tools should be a last resort. 10 When everything is straight, save the image again. If you are fitting several pieces, you might consider saving after you fit each piece. Though you can undo multiple times using the Histories, you do not want to lose your work to a crash. 11. If you have not done so, change the opacities of the Layers back to 100%. 12. Use a combination of erasure and blending to smooth the edges of the layers. If the images are placed well, and the scans match, blend the images with a simple edge erasure of the upper layer, using a soft brush (0% hardness). The diameter of the brush will vary depending on the dpi and size of the images you are knitting together. If the blends don't look quite right, or they appear too linear, look for a natural break; erase along image element contours if the size of the overlap permits. The first move in blending was to shrink the overlap to a reasonable working size as the two scans had too much overlap. The best place to blend this image was at the thinner part of the stem as there was simply less that could go wrong--anything that did would be easier to fix where there was less image to worry about. A quick selection with the Marquee tool and the globe and upper stem for the wineglass was deleted from the upper layer using Clear. After making the rest of the item opaque by clicking its layer and entering 50% in the opacity, a slight skew and distortion matched up the stems. In order to get the stems to blend well, a combination of erasing and cloning with the stamping tool was necessary. See the blending layers in Figure 03Cdfig01 on the CD-ROM. 13. When all edges have been smoothed to your satisfaction, flatten and save the image. How It Works Without changing the resolution or scanner settings, you can get similar tonality when scanning an image in parts. These parts can be knitted together in Photoshop to make one larger image. Targets help you align the two (or more) parts of an image for a better match, and scanning overlap gives you room to work with when blending. Putting together larger images this way can be involved, but care in your approach will get reasonably good results. Paying strict attention to placing targets, keeping images parallel to the scanner edges, and aligning the overlap areas on the scanner helps to reduce scanner distortion. Comments I once used this technique with 12 parts of a topographical map on a machine that was running 16MB of RAM and a 66mhz processor. The image was put together using a prelayer version of Photoshop (2.5). The resulting image was 250MB, which stretched the limits of the capacity of the hard drive (which was only 420MB) and its capability to use scratch disk space. That is to say, almost anything

file://I:\chapters\i\ix706.html

3/22/01

Adobe Photoshop 5 How-To - CH 3 - Scanning

Page 27 of 27

is possible considering the advantages offered by newer computers and the power of Photoshop's layers. Patience and care help get all the lines straight. Good scanning techniques and attention to detail get the raw material in scans so you can knit the parts of any image together. FIGURE 3.37 The gray area above represents the first scan, the white is the overlap. The image is altered this way to clarify the layers and is for demonstration purposes only. FIGURE 3.38 With a little tweaking (in this case, flattening the base to look like the glass was standing upright), the glass was finished and the parts matched very well.

file://I:\chapters\i\ix706.html

3/22/01

Information

27 pages

Find more like this

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate

635574


You might also be interested in

BETA
Microsoft Word - IJAMM1129proofed07.doc
-NDT_INDUST_RADIOGR_GB.qxd
Microsoft Word - v71-59.doc
HP Scanjet G3110 Photo Scanner