# 3470- Using Corel OCR-TRACE™ 8 This document will describe how bitmaps can be converted into vector file formats and scanned text can be converted to editable text. Converting Bitmap to Vectors Images can be opened using the File, Open command, or they can be scanned into Corel OCR- TRACE™ 8 through File, Acquire Image, then Acquire, or Acquire through CorelSCAN™ 8. Once the image is in Corel OCR-TRACE 8, a split window display will appear. The left side of this window contains the bitmap image and the right side will hold the converted vector image when finished. Depending on the type of image being converted, there are two main conversion options to choose from to give a good representation of the image, and four artistic methods used for special effects. To ensure a more detailed conversion, make sure that the images are created with a minimum resolution of 150 dpi or higher. For the best results, scan the images at 300 dpi. To perform the conversion, click on the OCR-Trace menu, select Perform Trace and a sub menu will appear. When the image is a line drawing with no fills, select "By Centerline": This method requires a black and white (1-bit) image, and converts the image to vector lines by tracing along the center the lines. When the image has filled in areas, select "By Outline": This method will convert an image by creating a series of vector objects. If it is necessary to convert photographs with high color depths, the image will need to be converted to 256 colors first. This can be done under the image menu and choosing convert to, then paletted (8-bit). Selecting an optimized palette with error diffusion will give the best results for the conversion. Artistic Tracing Woodcut method This method produces a vector graphic that contains objects of varying width, depending on the intensity of the original image at any given point. Sketch method This method produces a vector graphic that contains separate layers of lines, with each layer crossing at a different angle to create a mesh-like effect. Line spacing values can be set and are applicable to all layers. The Sketch method is mainly used for special effects. Mosaic method This method produces a vector graphic consisting of an array of symmetrical objects. Each object's color is based on the average color of the original image in that area. The number of tiles used can be changed, and, as the number increases, the original image becomes more recognizable. 3D Mosaic This method is very similar to the Mosaic method. In both these methods, the traced graphic is comprised of an array of tiled symmetrical objects. However, for the 3D Mosaic method, select the shape as well as the number of tiles. 3D Mosaic tiles look as if they are embossed. The Centerline, Woodcut, and Sketch tracing methods use intensity thresholds to determine which part of the bitmap image is traced. The intensity of a color refers to its pureness on a scale of 0 to 255, with 0 being black and 255 being the pure color. On a grayscale image, 255 would be white. The threshold is a value between 0 and 255 that determines when the tracing stops. For example, when setting a threshold of 128, any area in the original image that falls below 128 is traced, and any area above 128 is not traced. If the threshold were set at 0, only those areas that are pure black would be traced. If the threshold were set to 255, the entire image would be traced (since every part of the image has some intensity). Importing the Traced Image into CorelDRAW To import the converted image into CorelDRAW 8, click on the File menu in OCR Trace, select Save, and then Vector. Saving to the"Corel Presentation Exchange 6/7 (CMX)" file format is a good choice to allow further editing of the file in CorelDRAW 8. Trace Settings Different settings can be applied to achieve greater or less detail on the converted image. These settings are located on the property bar that runs across the top of the window. The settings listed here are dependant on the type of image, and the type of conversion being applied. To ensure the property bar is there, click on the View menu and look for a check mark beside "Property bar". If it does not have one, select it. A good trace can usually be achieved by selecting the Accurate preset on the "Select Preset" drop down list on the property bar. Below are some details on the additional settings that can be applied to enhance the conversion if needed. Node Reduction: Range 0-100. Altering this value will affect the amount of nodes that are used to create the vector line. The higher the value, the fewer nodes used. (Sometimes this setting will be needed to help pick up on different line thicknesses on a continuous line) Creating the image with less nodes, will also make it easier if the file needs to be edited afterwards. Noise Reduction: Range 0-30. This setting defines an area in pixels that will be ignored on the conversion. OCR-TRACE 8 Does not try to convert an area of pixels less than the value stated here. This setting is useful for getting rid of some background noise or added pixel information picked up on the scan. Iteration 1-255 : This sets the number of times OCR-TRACE 8 will try to reduce the lines in the image to determine the actual center of the line. Color tolerance: This setting allows for how much color depth carries through on the conversion. A low tolerance setting brings over all of the colors in the original bitmap, whereas a high tolerance reduces the amount of colors used in the conversion. Hole filling: This setting sets the accuracy of the hole filling applied to the conversion. Converting Bitmaps with Optical Character Recognition When a document is scanned into a digital format, a bitmap image of the page is produced. When the document contains text, the characters are treated as pictures and, as such, cannot be edited. The Optical Character Recognition (OCR) feature converts the bitmap image back into editable text so that the document can later be used in a text editing application. The scanned document should be Black and White or Grayscale and at least 200 dpi. The best OCR results, however, are obtained with a scanned image at 300 dpi. Once the image is open in Corel OCR-TRACE 8, click on the "OCR Trace" menu and select either: Perform OCR : This option will convert the image into editable text that will appear on the right hand side of the window. Perform OCR Trace: Select this option when the document has text and graphics. The finished conversion will have the editable text and also convert the graphic into a vector image. Occasionally, the whole document may not need to be converted. In situations such as this the "Draw OCR Block" option located on the tool bar can be used to define an area that the conversion will be applied to, while the rest of the document is ignored. Additional settings can be applied to the conversion by altering some of the settings on the property bar. Auto Verify: Enabling this feature will run the spell checker automatically on the converted text. OCR Confidence: This scale determines the level certainty of the character recognition. If it is not certain of the character it will replace it with the character specified in the box to the left of this setting on the property bar. (i.e. ~ ) Document Format: This drop down button gives the different layout options to retain on the conversion. Multi-column text, Single column text and Table formats. OCR Source: This drop-down list is needed to tell the application what source the document came from. Formatting: This setting determines additional formatting options for the text only.