-------------------------------- READRIGHT FOR WINDOWS DEMONSTRATION VERSION USER'S GUIDE --------------------------------- USING THIS GUIDE ---------------- This user's guide contains step-by-step instructions for scanning and converting documents or images to text with the demonstration version of ReadRight for Windows. This version allows you to do all of the procedures that you can perform with the full, off-the-shelf version of ReadRight, EXCEPT FOR SAVING. You may copy and distribute your demonstration disks as many times as you like. Before Using This Guide ----------------------- Before using this guide, you should know how to perform the following procedures in Microsoft Windows 3.x: * Use the mouse to select, click, drag, and choose commands from a menu * Use accelerator keys and shortcut keys * Make windows active or inactive * Open, size, move, close, and choose settings in windows * Open, close, and choose settings in dialog boxes * Change the path You may also find it helpful to be familiar with memory management in Microsoft Windows 3.x for the type of computer system you use. If you need to do so, refer to the MICROSOFT WINDOWS USER'S GUIDE for information. You should also know how to name and delete files and to create directories. You'll also need to know how to edit CONFIG.SYS and AUTOEXEC.BAT. These procedures are described in your DOS manual. SYSTEM AND MEMORY REQUIREMENTS ------------------------------ To use ReadRight, you must have * An IBM AT, PS/2, or 100% compatible. * Microsoft Windows 3.x. * At least 4MB of RAM, which must be configured at extended memory. * A hard disk. ReadRight uses about 5MB of disk space. You should also allow enough disk space to store the files you create when using the program. * One floppy disk drive: 5 1/4 inch (1.2MB) or 3 1/2 inch (720K). * MS-DOS or PC-DOS versions 3.1 or higher. * A mouse that's supported by Microsoft Windows 3.x. You can run ReadRight in both standard and enhanced modes, but the memory requirements will be different in each. Refer to the MICROSOFT WINDOWS USER'S GUIDE for detailed information on memory management in Microsoft Windows 3.x. Real Mode --------- You can't run ReadRight in real mode. Standard Mode ------------- If your computer has 4MB or more of RAM, ReadRight will run most quickly in standard mode. If you have less than 5MB of memory, SMARTDrive should be set to 1MB or less. 386 Enhanced Mode ----------------- If you're using a 386 computer (even one with less than 4 MB of RAM), you can run ReadRight in enhanced mode. Your permanent swap file should be set to 5 MB or more. Refer to the MICROSOFT WINDOWS USER'S GUIDE for detailed information on swap files. Checking Memory Status ---------------------- To see if you have enough memory to run ReadRight, follow these steps: 1 Start Microsoft Windows 3.x in standard mode. If you've installed ReadRight in enhanced mode, type the following line at a DOS prompt: win /s Then press Enter 2 Open the Help menu, and choose about Program Manager. You should have at least 2800 K of free memory. 3 Click OK or press Enter to close the dialog box. Freeing Memory -------------- If you don't have enough memory, look at your CONFIG.SYS file to see if it contains entries that you aren't using. If you're running memory-resident programs, you may need to delete some of them from AUTOEXEC.BAT. Modify CONFIG.SYS or AUTOEXEC.BAT with a word processor or text editor (save as an ASCII file), and then reboot your computer. INSTALLING READRIGHT -------------------- Installing Your Scanner Hardware and Software --------------------------------------------- Refer to the documentation that came with your scanner for instructions on installing your scanner and hardware and software. README.TXT may contain additional information on the hardware and software for your scanner. You can open and print README.TXT, which is saved in ASCII format, with your favorite word processor. Modifying CONFIG.SYS, AUTOEXEC.BAT, or WIN.INI ---------------------------------------------- Before installing ReadRight, make sure the files and buffers in CONFIG.SYS are set to AT LEAST the following: files=30 buffers=15 If you need to do so, you can use your word processor to modify CONFIG.SYS now. Be sure to retrieve and save the file as a DOS (ASCII) text file. CONFIG.SYS, AUTOEXEC.BAT, and/or WIN.INI must also contain information on your scanner before you can scan with ReadRight. Refer to README.TXT for this information. Be sure to reboot your computer after making the modifications to these files. Installing the Demo Disk -------------------------- Before you install ReadRight, make sure Microsoft Windows 3.x is installed on your computer. This procedure is explained in the MICROSOFT WINDOWS USER'S GUIDE. The following procedure installs ReadRight on C:\OCR, unless you choose another directory. Use this procedure after you've installed and tested your scanner hardware and software. To install ReadRight, follow these steps: 1 Place Disk 1 in disk drive A, and close the door. You can install from any floppy drive, although drive A is the most common choice. 2 Type: A: and press Enter 3 Type: install The installation procedure begins 4 Follow the instructions on the screen to install ReadRight onto your hard disk. When installation has completed, the program displays this message. Installation Successful Press Any Key Be sure to read C:\OCR\README.TXT If you chose a directory other than C:\OCR during installa- tion, README.TXT is located in that directory. You can read README.TXT by opening it in your word processor or other editor. 5 Press any key on your keyboard. Placing ReadRight in a Program Group ------------------------------------ Before beginning to use ReadRight, you might want to make ReadRight a program item in one of your program groups in Microsoft Windows 3.x. This way, you can start ReadRight quickly by double-clicking on the ReadRight icon. After choosing NEW from the File menu (in Microsoft Windows 3.x), type ReadRight in the DESCRIPTION box, and type C:\OCR\RR.EXE in the COMMAND LINE box. (If you chose another directory during installation, type the name of that directory, rather than c:\ocr.) When you click OK, the ReadRight icon appears in the active program group. STARTING READRIGHT ------------------ To start ReadRight from Microsoft Windows 3.x, follow these steps: 1 Turn on your computer and follow your usual procedure for starting Microsoft Windows 3.x. 2 Open the Program Group that contains ReadRight. The ReadRight icon is located in the program group in which you placed it after you installed ReadRight. 3 Double-click the ReadRight icon. You can also start ReadRight from the Microsoft Windows File menu. Follow these steps: 1 Open the File menu, and choose Run. 2 Type the path for ReadRight. If you installed ReadRight on the default path, type c:\ocr\rr.exe. 3 Click OK or press Enter. A QUICK TOUR OF THE READRIGHT INTERFACE --------------------------- When ReadRight has loaded, the ReadRight user interface appears. As an application that runs under Microsoft Windows 3.x, ReadRight is designed around a structure of pull-down menus, windows, and dialog boxes, which you can easily access with the click of a mouse button or a couple of keystrokes. The Image Window ---------------- The Image window, which is located on the left side of your computer screen, is always open when you're running ReadRight. In this window, the bitmapped image appears whenever you scan a page or open an image file. Once an image appears in this window, you can convert the words on the image to text with ReadRight. This process is called recognition. The Menus --------- Although you can start scanning and recognition from the Image window, you'll use menu commands to customize recognition results. The menu bar contains six menus: * FILE, which lists commands for opening and saving image files, saving and closing text documents, and loading and saving settings files, and exiting ReadRight. * EDIT, which lists commands for removing a page of recognized text from and checking for recognition errors in the text document. * RECOGNITION, which lists commands for scanning, recognizing, and changing between page and stack modes. * IMAGE, which lists commands for changing the image magnification, rotating the scanned image, locating text on the image, using a template on the image during recognition, defining the content and format of particular regions, and showing the order that regions are linked. * SETUP, which lists commands for selecting the scanner model that you're using, previewing the scanned image, selecting page parameters, and choosing typographic and formatting characteristics for the text document. * HELP, which lists commands for accessing online help and for opening the About ReadRight box. You can use the online help for additional infor- mation on using ReadRight after you've gone through this user's guide. Buttons for Scanning and Recognizing ------------------------------------ The Image window contains three buttons that you can use to start ReadRight's major processes: scanning and recognizing. When you want to scan and recognize a page in one automatic process, you'll click the SCAN AND RECOGNIZE button. And you'll chose the SCAN and RECOGNIZE buttons separately when you want to scan a page first and recognize it after you've taken a look at how well it has scanned. The Toolbox ----------- ReadRight's toolbox is located along the left side of the Image window. The toolbox contains these tools: * SELECT TOOL, which you'll use to select a region that's on the image in the Image window. A region is an area on the image that you want ReadRight to process in some way--usually, an area containing text that you want ReadRight to recognize. Before you can resize, delete, or otherwise modify a region, you have to select it. * MAGNIFY TOOL, which you'll use to enlarge the size of the image in the Image window so that you can look more closely at specific areas. * COMPOUND TOOL, which you'll use to define (create) compound regions. Compound regions contain text and graphics and/or more than one column of text. * TEXT TOOL, which you'll use to define text regions. Text regions are made up of single columns of text. In a process called page decomposition, ReadRight breaks compound regions into text regions before recognizing the text. * GRAPHICS TOOL, which you'll use to define graphics regions. You can define graphics regions when you want to save the line art on your page as TIFF or PCX image files. * BLANK TOOL, which you'll use to "blank out" an area that you don't want ReadRight to recognize. (These areas are called blank regions.) Sometimes it's easier to blank one or two areas that you don't want than it is to define all of the areas that you do want. * LINK TOOL, which you'll use when you want ReadRight to recognize the text regions on the image in a particular order. Linking is useful when you want to change the order of paragraphs in the text document (the output document produced by ReadRight that contains the recognized text). At the bottom of the toolbox are these two icons: * PAGE ICON, which you'll click when you want to scan and/or recognize a page at a time (a procedure called page mode or page processing). * STACK ICON, which you'll click when you want to scan and/or recognize long documents or many pages at a time (a procedure called stack mode or stack processing). In stack mode, your documents can consist of single- or double-sided pages. GETTING READY TO SCAN AND/OR RECOGNIZE -------------------------------------- In this section, you'll learn to choose a scanner model and page settings and to test your first scan. Locate a simple page, such as a memo or letter, for this first exercise. You can try more complex pages later. Choosing a Scanner ------------------ The first time you use ReadRight, you must choose the scanner model that you've installed on your computer so that you can scan. Follow these steps: 1 Open the Setup menu, and choose SELECT SCANNER. The Select Scanner window appears. 2 Scroll through the select box to find the scanner model that you've installed, and click it. If you're using a nonsupported scanner or if you aren't using a scanner with ReadRight, choose NO SCANNER in the select box. 3 Click OK. When you click OK, ReadRight checks whether the scanner is active. If the scanner isn't on or isn't installed, an error message appears. You must turn on or install the scanner, and then reselect the scanner model in SELECT SCANNER before you can scan with ReadRight. Choosing Default Page Settings ------------------------------ NOTE: If you're not using a scanner or if you're using a nonsupported scanner, follow the instructions for "Choosing Default Page Settings" and then proceed to "Recognizing Page by Page." ReadRight contains many commands and settings that you can use to customize recognition results and to make scanning and recognizing occur more quickly and accurately for the types of documents that you typically use. One of these commands is PAGE INFO, with which you can customize certain settings for the page that you'll scan and recognize with ReadRight. Follow these steps: 1 Open the Setup menu, and choose PAGE INFO. The Page Info window appears. 2 Use the default CONTENT (COMPOUND). The icon for Compound looks like two boxes, each containing the letter T. When you first scan a document, ReadRight identifies the entire scanned image as a single region. ReadRight uses the Content setting to determine what kind of information is in this region. If the CONTENT is COMPOUND, before recognition, ReadRight automatically runs a process called page decomposition, which locates all of the text on the image and divides the text into individual text regions. If you scan a variety of documents, you should use Compound as the page content. 3 Select a unit of INCHES or CENTIMETERS. The UNITS you choose indicate whether you can to measure the scanned image with an English or metric ruler. INCHES is the default unit of measure. If you'd prefer to use a metric ruler, click CENTIMETERS. 4 Use the default ORIENTATION of portrait text. ORIENTATION indicates the direction of the text on the page (landscape or portrait text). ReadRight uses the orientation you choose to determine whether and how to rotate the image so that you'll be able to read the text on the image in the Image window. 5 Close the Page Info window. Testing Your Scan ----------------- The first step in page recognition with ReadRight is to scan the page. Scanning converts the printed page into a bitmapped image that ReadRight uses during recognition. Before beginning to scan, you should check the settings for your scanner to make sure they're appropriate for the page that you're scanning. With SCANNER SETTINGS on the Setup menu, you can test how your scanner converts the page to an image before you begin recogni- tion. Scanner settings include options like resolution, page size, and brightness. All of these settings must be adjusted specifically for the page you're scanning. Otherwise, you're likely to get poor recognition results. In this exercise, you'll learn to test the scanner settings for the sample page that you've selected. Follow these steps: 1 Place the sample page in portrait orientation on your scanner. 2 Open the Setup menu, and choose SCANNER SETTINGS. The Scanner Settings window appears. 3 Select a PAPER SIZE. Choose LETTER or LEGAL, depending upon the size of your sample page. The Page Image area (the area on the right side of the window that's bordered by the ruler) adjusts to suggest the page on the actual scanner plate. 4 Select a RESOLUTION. RESOLUTION refers to the clarity of detail in the image produced by the scanner. The resolution you select depends upon the size of the text on the page you're scanning. 300 is the usually best setting unless you're scanning very small text (less than 8 points). 5 Adjust BRIGHTNESS to about 70%. Sometimes the page you're scanning is so light or dark that the text is broken or merged, which could cause recognition problems. You can't change the physical page, at least not without a great deal of work. But by adjusting BRIGHTNESS, you can improve recognition by altering the lightness or darkness of the scanned image. ReadRight measures BRIGHTNESS in percentages (0 to 100%). Moving BRIGHTNESS toward 0% lightens the image, while moving BRIGHTNESS toward 100% darkens the image. The available percentages depend upon those available on the scanner. Some scanners have many brightness steps, while others have only a few. 6 Click PREVIEW. When you click this button, ReadRight scans the page at the lowest resolution available on your scanner. An image appears in the Page Image area, and you see the miniscan box (the small white square in the middle of the Page Image area) on the image. Once you preview the page, you can check the overall clarity of the image. You can also check the placement of the miniscan box and move the miniscan box if necessary. You should always miniscan specific areas of the previewed image to be sure the image is clear enough for recognition. 7 If necessary, drag the miniscan box to another area on the image. Generally, you'll move the miniscan box to an area containing text that looks fairly clear. If you move the miniscan box to an area where you think recognition problems might occur (such as a very dark or very light area), you might lighten or darken the image too much when you adjust BRIGHTNESS, and recognition problems could occur. Make sure the area inside the miniscan box contains text. 8 Click MINISCAN, and check the clarity of the magnified image. Click this button to rescan the area inside the miniscan box. After scanning, this area is magnified dot-for-dot in the Miniscan Image area. The magnified image should show characters that are clear and well defined. The characters shouldn't be broken or merged. Usually, the only scanner settings that you'll need to adjust after scanning is BRIGHTNESS. Look at the image in the miniscan box. If the image seems at all broken, increase BRIGHTNESS slightly. Then repeat steps 7 and 8. When you're satisfied with the clarity of the image in the Miniscan Image area, close SCANNER SETTINGS. For Further Information ----------------------- For further information on the commands you've covered in this section, read about the following information in ReadRight's online help: Page Info Regions Scanner Settings Select Scanner You may find it helpful to at least skim this information before you begin the next section. To access the online help, follow these steps: 1 Press F1 This step opens the Index help window. 2 Scroll through the list of topics to find the topic you want to read. 3 Move the cursor to the topic. Notice that the cursor changes to a hand when you move the cursor over the topic. 4 Click the topic. 5 An explanation of the topic appears. After reading the explanation, click the Index button to return to the list of ReadRight topics. Your MICROSOFT WINDOWS USER'S GUIDE contains detailed information on using online help. RECOGNIZING PAGE BY PAGE ------------------------ In this section, you'll learn to scan and recognize individual pages in two ways: automatically and interactively. You'll use automatic scanning and recognition when you want these two processs to occur in one quick procedure. If you want to look at the scanned image before recognition or if you want to identify which parts of the scanned image ReadRight should process, you'll use the interactive procedure. You'll also learn to open an image file, if you're using this demonstration version of ReadRight without a supported scanner. Working in Page Mode -------------------- When you're working with pages one at a time, ReadRight must be in Page mode (sometimes called page processing). Page mode is the default condition for scanning and recognizing with ReadRight. You can choose Page mode with one of these steps: * Click the Page icon on the Image window. The page icon is highlighted when it's selected. or * Open the Recognition menu, and choose PAGE. A check mark appears to the left of PAGE when it's selected. You can use the same sample page that you used earlier, or you can use the image file, SAMPLE.TIF, which is on Disk 2 of the ReadRight for Windows demonstration disks. Recognizing Automatically ------------------------- Before you can recognize a page, an image of the image must appear in the Image window. Choose SCAN AND RECOGNIZE when you want to scan and recognize a page in one automatic procedure. You'll choose this command when you don't need to look at the scanned image before recognition. Follow these steps: * Click the SCAN AND RECOGNIZE button on the Image window. or * Press Ctrl + B or * Open the Recognition menu, and choose SCAN AND RECOGNIZE. ReadRight scans the page, and the image of the page appears in the Image window. This image is sometimes referred to as the scanned image or the on-screen image. After scanning, ReadRight automatically begins a process called page decomposition (since, earlier, you chose COMPOUND as the CONTENT in PAGE INFO). During page decomposition, ReadRight locates all of the the text on the page and breaks the text into text regions. THese regions are rectangular or polygonal outlines that identify the ares of text which ReadRight will recognize. After page decomposition, ReadRight immediately begins to recog- nize the page. The scanned image temporarily disappears from the Image window, and gray blocks show which text regions ReadRight will recognize. The Document window appears next to the Image window, and the recognition feedback box appears. ReadRight shows the recognized text in this box so that you can monitor the progress of recognition to make sure no problems occur. When recognition is complete, ReadRight beeps, the feedback box disappears, and ReadRight places the recognized text in the Document window. NOTE: If recognition is unsatisfactory, try increasing BRIGHTNESS in SCANNER SETTINGS (Setup menu), and then choose SCAN AND RECOGNIZE again. Checking for Errors ------------------- Once you've recognized the text, you should check for recognition errors. Although you can use your word processor to check for errors, spelling checkers only identify misspelled words. Read- Right's error checker, however, finds all of these types of words: * UNRECOGNIZED. Words containing single characters that Read- Right can't identify with a high degree of certainty. * QUESTIONABLE. Words that ReadRight has recognized but that you should double-check. * MISSPELLED. Words that aren't in ReadRight's built-in or user dictionaries. If you check for misspelled words, you can also select IGNORE WORDS IN ALL CAPS so that ReadRight won't check words that consist entirely of capital letters (such as acronyms), since these words often aren't in dictionaries. NOTE: Unrecognized characters and Questionable words are marked with characters that you can choose in OUTPUT OPTIONS (Setup menu). For now, use the default charac- ters that are already identified. Follow these steps: 1 Press Ctrl + E or Open the Edit menu, and choose ERROR CHECK. The Error Check window appears. 2 Identify the types of words that ReadRight should find. For this exercise, select all of the settings: UNRECOG- NIZED, QUESTIONABLE, MISSPELLED, and IGNORE WORDS IN ALL CAPS. These settings may already be selected. 3 Click START CHECK. An image of the first word appears in the WORD IMAGE box. The unrecognized word also appears as the WORD IN QUESTION at the top of the window and in the CHANGE TO box. 4 Compare the WORD IN QUESTION against the image. Decide whether the word is correct or, if the word contains unrecognized characters, decide how to correct the word. 5 If the WORD IN QUESTION is correct, follow one of these steps: * Click NO CHANGE. Another WORD IN QUESTION appears. Begin again with step 4. or * Click ADD TO DICTIONARY. Follow this step if the word is correct and you want to add it to your user dic- tionary. If you do this, each time ReadRight encounters the word, it recognizes the word as legitimate. 6 If the word isn't correct, follow one of these steps: * Click SUGGEST for suggestions on how to correct the word. A list of suggestions appears in the SUGGESTIONS box. If one of the suggestions is correct, click the word, and then click CHANGE. The suggested word appears in the CHANGE TO box and replaces the incorrect word in the text document. or * If necessary, edit the word in the CHANGE TO box. Then click CHANGE. The corrected word replaces the WORD IN QUESTION and the word in the text document. When ReadRight has checked the text document for the types of word you chose, a message states that all errors have been checked. 7 Click CANCEL to close ERROR CHECK. Recogizing Interactively ----------------------- Sometimes, before recognition, you'll want to look at the scanned image in the Image window, rather than in SCANNER SETTINGS. You also may want to identify particular regions that you want ReadRight to recognize, or you may even want different regions to be recognized in separate text docu- ments. In instances like these, you should scan and recognize in two separate processes. Before you can recognize a page, the image of the page must be in the Image window. You can place an image in the Image window through one of two procedure: * By scanning the image with ReadRight and your scanner. * By scanning an image with image-scanning software, saving it in TIFF or PCX format, and opening the image with ReadRight. This procedure is especially useful if you don't have a supported scanner. You'll learn both of these procedures in this section. Scanning the Page (Nonsupported Scanners or No Scanner) ------------------------------------------------------ ReadRight doesn't control the scanning process with non- supported scanners. If you aren't using a supported scanner, you must use the image-scanning software that came with your scanner to scan the page. Refer to the documentation for the software for instructions. Be sure to save the image in TIFF or PCX format. NOTE: If you don't have a scanner connected to your computer system right now, you can use the image, SAMPLE.TIF, which was copied to your hard disk during installation. You must, however, have a full-page scanner if you decide to purchase ReadRight. Use the following instructions to open SAMPLE.TIF. After you scan the page and save it as an image file, you can open the image in the Image window. Follow this step: 1 Open the File menu, and chose Open Image File. If an image appears in the Image window at the wrong orientation, you can rotate it by choosing ROTATE from the Image window. Once the image appears at the correct orien- tation, you can begin recognition. Notice the compound icon on the top left side of the scanned image. This icon, which looks like the compound tool, represents the COMPOUND page content that's selected in PAGE INFO. This icon also appeared on the image when you scanned and recog- nized earlier, but you might not have noticed it then, since page decomposition and recognition began right away. Proceed to the section, "Defining Regions." Scanning the Page (Supported Scanners) ------------------------------------- To scan the sample page you're using, follow one of these steps: * Click the SCAN button on the Image window. or * Press Ctrl + A * Open the Recognition menu, and choose SCAN. ReadRight scans the page, and the image appears in the image window. But this time, page decomposition and recognition won't automatically occur. When you're working interactively, these processes will occur only when you ask for them. Notice the compound icon on the top left side of the scanned image. This icon, which looks like the compound tool, represents the COMPOUND page content that's selected in PAGE INFO. This icon also appeared on the image when you scanned and recog- nized earlier, but you might not have noticed it then, since page decomposition and recognition began right away. Defining Regions ---------------- During automatic scanning and recognition, you saw Read- Right define regions itself. When ReadRight defines regions automatically, all areas of text on the page are separated into text regions. (You can run this process yourself by choosing FIND TEXT REGIONS from the Image menu.) But you can create your own regions when you want to control which parts of the image ReadRight should recognize. Only the text information inside the regions you define will be included in recognition. To define a text region, follow these steps: 1 Click the text tool. The text tool, which looks like a T in a box, is located in the toolbox on the left side of the Image window. When you click, the text tool becomes highlighted, and the cursor changes to a crosshair. 2 Move the cursor to the top left corner of the area of text that you want to enclose in a region. Define a text region in an area that contains only single columns of text. NOTE: Text regions can ONLY consist of single columns of text. If the area you enclose in a region contains more than one column of text and/or both text and graphics, you must define a compound region so that ReadRight runs page decomposition on the region to locate the text. 3 Press and hold the mouse button, and drag the crosshair diagonally to the bottom right corner of the text area. The region, which looks like a rectangular box, appears as you drag the cursor. 4 When the region encloses the column, release the mouse button. The region has handles (small square boxes) in its corners and along its sides, indicating that the region is selected. An icon of the text tool appears in the top left side of the region. This icon is called the region type label. Notice that the compound icon has disappeared from the top left side of the scanned image. Since you've defined a text region, ReadRight no longer assumes that you want to run page decomposition on the image, so the compound icon disappears. 5 Open the Image menu, and choose REGION CONTENT. Notice that TEXT is selected for the region you just defined. With REGION CONTENT, you can change the region type without deleting and redefining it, and you can indicate the content or format of a region so that ReadRight can recognize it with greater accuracy. 6 Click OK or press Enter. The REGION CONTENT dialog box closes. 7 Click the select tool. The cursor changes to a pointer, and the handles disappear from the text region. When the handles aren't visible, the region is unselected. After you've defined a region, it's best to unselect the region so that you don't inadvertently change it. Examining the Image "Up Close" ------------------------------ After you define a region, you'll sometimes want to magnify the image where the region is located to make sure you enclosed all of the text that you want ReadRight to recog- nize. You can magnify the entire image by choosing DOT FOR DOT VIEW or MAGNIFIED VIEW from the Image menu. (The image is at REDUCED VIEW when you first scan or open it.) But if you want to magnify a particular area on the image, you should use the magnify tool. Follow these steps to magnify the image around the area you just defined: 1 Click the magnify tool. The magnify tool, which looks like a magnifying glass, is located in the toolbox. When you click the tool, the cursor changes to a magnifying glass. 2 Move the cursor inside the region you just defined. 3 Click the mouse button. The image enlarges to MAGNIFIED VIEW. If you click again, the image changes to DOT FOR DOT VIEW. If you can't see all of the region, use the scroll bars on the right side and bottom of the Image window to move around the image. After you magnify the image, you may decide that you want to resize the region that you defined. First select the region by clicking the select tool and then clicking inside the region label. Then move the pointer to the handle on the side of the region that you want to resize. Press and hold the mouse button, and drag the handle until the region is the appropriate size. Then click the select tool to unselect the region. 4 Open the Image menu, and choose REDUCED VIEW. The scanned image changes to its original size. To learn more about regions, refer to the topic, "Regions," in the online help. Recognizing the Regions ----------------------- Now that you've defined the regions that you want ReadRight to recognize, you can begin recognition. Follow one of these steps: * Click the RECOGNIZE button on the Image window. or * Press Ctrl + R or * Open the Recognition menu, and choose RECOGNIZE. ReadRight first breaks the compound region into text regions, and then begin recognition. The image temporarily disappears, and the text regions that ReadRight will recognize are replaced with gray blocks. The recognized text appears in the recognition feedback box, and, when recognition is complete, ReadRight beeps and then places the text in the Document window. Scanning and Recognizing a Second Page -------------------------------------- Before scanning a second page (or opening another image file), you should check for errors on the first page. When you scan again, the scanned image of the first page in the Image window is replaced with the image of the second page. Since ERROR CHECK uses the image in the Image window to show you the bitmapped WORD IMAGE, if you don't check for errors on the first page before recognizing again, you won't be able to see the bitmapped image of the WORD IN QUESTION. Follow the same procedure that you learned earlier to check for errors. Then follow this step to scan the second page: NOTE: If you're using SAMPLE.TIF, you don't need to open the image again, since it's still in the Image window. Follow the instructions for recognizing later in this section. 1 Click SCAN on the Image window (or follow one of the other procedures that start scanning). ReadRight scans the page on your scanner. To recognize the page, follow this step: 1 Click RECOGNIZE on the Image window (or use one of the other procedures that starts recognition.) Page decomposition occurs, and then recognition begins right away. ReadRight beeps when recognition is complete. 2 Use the page arrows and scroll bars on the Document window to look at the text document. The bottom left side of the Document window contains two arrows. Click the up arrow to move to the first page of the text document, and click the down arrow to move back to the second page. Maximize or resize the window, or use the scroll bars at the bottom and right sides of the Document window to look through a page of text. Once ReadRight has recognized the text, check for recognition errors on the second page of the text document (ERROR CHECK on the Edit menu). Make sure the second page of text is in the Document window before starting ERROR CHECK. For Further Information ----------------------- For further information on the commands and tools you covered in this chapter, refer to these topics in ReadRight's online help: Dot for Dot View Error Check Find Text Regions Magnified View Open Image File Output Options Page Recognize Reduced View Region Content Rotate Scan Scan and Recognize For related information that will help you to become more skilled in Page mode recognition, refer to these topics: New Linking Regions Regions You may filed it helpful to at least skim these topics before moving on to the next topic. RECOGNIZING STACKS OF PAGES --------------------------- In this section, you'll learn to scan and recognize stacks of pages and image files. You'll use stack processing when you want scanning and recognition to occur in one automatic procedure. If you want to first scan pages into image files and later recognize them with ReadRight when you aren't using your computer, you'll use deferred processing. You'll also learn about templates and settings files in this chapter. Templates and settings files will help you to process similar documents more quickly. Before beginning this chapter, you should be familiar with the commands and processes discussed earlier. NOTES: * If you plan to do a lot of stack processing, it's highly recommended that you use an automatic document feeder (ADF) with your scanner. * If you don't have a scanner connected to your computer system, right now, you'll only be able to do some of the procedures in this section. Before continuing, make one or two copies of SAMPLE.TIF, follow the instructions for "Working in Stack Mode," "Creating Templates and Settings Files," and skim the rest of the section to "Deferred Processing." Follow the instructions for recognition under "Deferred Processing." Working in Stack Mode --------------------- When you want to scan and recognize longer documents, Read- Right should be in Stack mode. Although you can process long documents in Page mode, you'll process these documents much more quickly if you work with them in stacks. A stack is a group of pages--often belonging to the same printed document-- that you want to scan and recognize into the same text document. You can change to Stack mode through one of these steps: * Click the STACK icon on the Image window. The Stack icon is highlighted when it's selected. or * Open the Recognition menu, and choose STACK. A check mark appears to the left of STACK when it's selected. Before starting stack scanning, find two or three pages with these characteristics: * They should have clear, well-defined text. If you aren't using very clear documents, read "Getting the Best Results" in the online help before proceeding. * They can contain text and/or graphics. * They should all be the same size (English, letter or legal; metric, A4 or B5) and have the same orientation (portrait or landscape). * They can be single-sided or double-sided. (For now, don't use any blank pages or double-sided pages with blank sides. Blank pages signal ReadRight to stop the scan, although a message will ask you if this is really what you want to do.) * If possible, they should have the same layout. Once you've gathered these pages, you can begin the exercises in this chapter. Creating Templates and Settings Files ------------------------------------- If you commonly scan and recognize documents in the same way or for the same purpose, you may want to design settings files for these documents. Settings files contain groups of settings that are appropriate for scanning and/or recognizing a particu- lar type of document. You don't have to make settings files to use ReadRight effectively, since ReadRight always "remembers" the last settings that you used. But if you design settings files for the types of documents that you scan and recognize many times, you won't have to remember what settings work best each time you process these documents. Settings files can consist of this information: * SCANNER SETTINGS * OUTPUT OPTIONS * PAGE INFO * FORMAT (SAVE DOCUMENT AS and/or SAVE IMAGE AS) NOTE: Since saving a file is disabled in the demonstration disks, you won't be able to save format in your settings file. * Templates (the region information for the page), which are optional NOTE: Although you're learning about settings files in this section on stack recognition, you can use settings files in Page mode as well. Making a Settings File ---------------------- For this exercise, you'll learn to create a settings file for the pages in your stack. Follow these steps: 1 Adjust the settings in SCANNER SETTINGS (Setup menu) for your stack. Use one of the pages in your stack to test the scanner settings. Make sure you test BRIGHTNESS, even if you're certain the other settings are appropriate for the page. If you're using the ADF on your scanner, select USE ADF FOR STACK SCANNING. 2 Adjust the settings in OUTPUT OPTIONS (Setup menu) for your stack. Choose the options that you'd like to retain in your text document. Remember that ReadRight may not use all of the OUTPUT OPTIONS that you've selected. ReadRight only uses the options that are supported by the FORMAT you selected when you save (SAVE DOCUMENT AS). 3 Adjust the settings in PAGE INFO (Setup menu) for your stack. Choose the CONTENT, UNITS, and ORIENTATION for the pages in the stack. Remember to choose TEXT as the page CONTENT for single- column, text-only pages. If the pages in the stack have different page layouts or if the pages contain columns or both text and graphics, select COMPOUND as your CONTENT so that ReadRight will run page decomposition on each page before recognition begins. 4 If all of the pages in the stack have the same layout, design a template for your stack. Templates consist of the region information for the page: the regions defined in the Image window, the REGION CONTENT for each region, the order that regions are linked, and the type of region label that's showing when you save the settings file. If you define a template for pages with the same layout, ReadRight will use this region information to process each page in the stack. For example, if you define a text region on the upper left side of the template, ReadRight will try to recognize text on the upper left side of each page. However, if one of the pages has a graphic in this area, ReadRight will still try to recog- nize the graphic as text, since it's recognizing according to the layout of the template. For this reason, you should only use a template if all of the pages have the same layout. 5 Open the File menu, and choose SAVE SETTINGS AS. a Type a name for the settings file. For this exercise, use the name, FIRSTSET.SET. The extension .SET is automatically added to the filename if you don't type an extension. b Choose the directory where you'd like to save the settings file. c Click SAVE TEMPLATE if you designed a template for the page. If you haven't defined any regions on the image, this setting is dimmed. d Click Save to save the settings file. When you save a settings file, the settings remain active until you change them. Any changes to settings that you make after saving the file won't automatically become part of the settings file. To update a settings file, you must overwrite the save settings file with the changed version. Stack Processing ---------------- With stack processing, you can scan and recognize a stack of pages into a single, multipage text document. If you're scanning and recognizing a large stack of documents, it's a good idea to scan and recognize the first page in Page mode to make sure the settings you've chosen are correct. Since page condition can vary greatly, if you've loaded a settings file, you might want to at least check BRIGHTNESS (SCANNER SETTINGS) to make sure this setting is appropriate for the particular stack you're scanning. NOTE: If you're scanning double-sided pages and you test the page in Page mode, be sure to scan and recognize both sides of the page before changing to Stack mode. ReadRight begins to count the pages (to place them in the proper order) from the time you change to Stack mode. If you don't start with a new front page when you begin Stack scanning, the pages won't be sorted properly in the text document. Since you're only scanning a few pages, you don't need to test recognition now, but keep these ideas in mind. To process your stack, follow these steps: 1 If you designed a template in your settings file, open the Image menu, and make sure Use Current Template has a check mark beside it. When USE CURRENT TEMPLATE is checked, the template in the Image window is used for all pages in the stack. If USE CURRENT TEMPLATE isn't checked, the template is used for only the first page and disappears from the Image window when the next page is scanned. (ReadRight then uses the CONTENT in PAGE INFO to process the page.) * If USE CURRENT TEMPLATE is checked, close the Image menu. * If USE CURRENT TEMPLATE isn't checked, click the command. USE CURRENT TEMPLATE becomes active. 2 Place the stack on your ADF, or position the first page of the stack on your scanner. 3 Click the SCAN AND RECOGNIZE button on the Image window. or Press Ctrl + B or Open the Recognition menu, and choose SCAN AND RECOGNIZE. The Scan and Recognize Stack dialog box appears. 4 Choose SINGLE-SIDED PAGES if the pages have information on only one side, or choose DOUBLE-SIDED PAGES if the pages have information on both sides. 5 Click SCAN. If you're using an ADF, ReadRight automatically scans and recognizes each page in the stack: * If the pages in your stack are double-sided, when the last page is completed, ReadRight beeps to signal that recognition is complete. * If the pages in your stack are single-sided, after the odd pages are scanned and recognized, a message directs you to turn over the stack. Position the second side of the stack on the scanner, and click OK. The even pages are scanned and recognized into the same text document as the previous pages. The pages are placed in the proper order as the pages are scanned and recognized. If you're not using an ADF, you'll need to place each new page on the scanner. * If the pages in your stack are single-sided, after each page is scanned and recognized, a message directs you to place a new page on the scanner or to end the process. Click OK after placing the new page on the scanner. Continue this process until ReadRight has scanned and recognized all of the pages. * If you're scanning double-sided pages, after the first side is scanned and recognized, a message directs you to turn over the page. Click OK after turning over the page, and ReadRight scans and recognizes the other side of the page. Continue this process until ReadRight has scanned and recognized all of the pages. After the pages are recognized, you should check for recognition errors in the text document (ERROR CHECK on the Edit menu). When you check for errors in a stack of pages, the WORD IMAGE only appears when you're checking the last page, although ReadRight will show the WORD IN QUESTION for all pages. During ERROR CHECK, ReadRight uses the image in the Image window to show the WORD IMAGE. Deferred Processing ------------------- Sometimes, you don't have time to recognize an entire stack at one time, often because you need your computer to do other things. With ReadRight, you can scan a stack of pages into separate image files, and you can later recognize these image files when you aren't busy with your computer. This procedure is called deferred processing. The first stage of deferred processing is to scan the pages. ReadRight scans each page in the stack and saves it as a TIFF or PCX image file. NOTE: For this exercise, you should have about 1 MB of free disk space for each page/side in the stack. The actual disk space requirements vary with the image file format you chose. Follow these steps: 1 Place the stack on your ADF, or position the first page of the stack on your scanner. Use the same pages that you used in the last exercise. 2 Click the SCAN button on the Image window. or Press Ctrl + A or Open the Recognition menu, and choose SCAN. A dialog box appears. 3 Change to the directory where ReadRight should save the image files. For now, you can save the images in the current directory. 4 Choose a FORMAT for the Image files. You can save images in any of the following formats: PCX, TIFF Group 3 Compression, TIFF PackBits Compression, and TIFF Uncompressed. 5 In the FILENAME box, type PAGE.TIF or PAGE.PCX as the name for the image files. Type PAGE.TIF if you choose a TIFF format in step 4, and type PAGE.PCX if you chose PCX format. ReadRight sequentially adds a number to the prefix of the filename (in this exercise, the prefix is PAGE) to indicate the order in which the pages are scanned. For example, since the prefix is PAGE, the prefix of the first page becomes PAGE1, of the second PAGE2, and so on. NOTE: Remember that DOS restricts filename prefixes to eight characters. If you're scanning a document with less than 10 pages, limit the prefix to seven charac- ters. If the document has 10 to 99 pages, limit the prefix to six characters. 6 Choose SINGLE-SIDED if all pages have information on only one side, or choose DOUBLE-SIDED if the pages have informa- tion on both sides. 7 Click SCAN. ReadRight will display the same messages that you saw during stack scanning. Follow the instructions in these messages. If you're scanning double-sided pages with an ADF, ReadRight will number the pages on the first side of the stack (usually, the odd pages) as they're scanned, and then, when you turn over the stack, will number the pages on the second side (usually, the even pages). For example, if you scan a 10-page report that you name REPT.TIF, the odd pages are named REPT1.TIF, REPT3.TIF, and so forth to REPT9.TIF. THen, when you scan the even pages, ReadRight will name them REPT10.TIF, REPT8.TIF, and REPT2.TIF. (Notice that ReadRight numbers from the end of the stack when you turn it over. To begin recognition, follow these steps: 1 Click the RECOGNIZE button on the Image window. or Press Ctrl + R or Open the Recognition menu, and choose RECOGNIZE. The Recognize Stacks dialog box appears. 2 If necessary, change to the directory where you saved the image files. 3 Click PAGE1.TIF or PAGE1.PCX in the select box. Choose the name that you typed when you scanned the image file. If you're using the sample image that came with the ReadRight demonstration disks, choose SAMPLE.TIF. 4 Click ADD. The selected filename now appears in the Selected Files box. ReadRight recognizes images in the order they appear in this box. 5 Repeat steps 3 and 4 until all of the images that you want to recognize appear in the Selected Files box. Choose the images in the order that you want them to appear in the text document. Since the image file names were numbered sequentially when you scanned, if you choose the files in numeric order now, they should be in the correct order for recognition. If you scanned double-sided pages on the ADF, the files were sorted in the proper order as they were scanned and saved. 6 Click RECOGNIZE. Each image file is opened and recognized into the same text document of multiple pages. After the image files are recognized, check for errors in the text document (ERROR CHECK on the Edit menu). For Further Information ----------------------- For further information on the commands and tools that you covered i this section, refer to these topics in ReadRight's online help: Load Settings Recognize Save Settings As Scan Scan and Recognize Stack Templates Use Current Template CONGRATULATIONS! ---------------- You've now had your first look at the benefits of using ReadRight for Windows! Refer to the brochure that was enclosed with your ReadRight demonstration disks for further information on ReadRight, and feel free to call the Sales Representatives at OCR Systems if you have any questions (1-800-233-4627).