Autodesk 05726-091452-9060 User Guide - Page 104

WiseImage - User's Guide 3. Enter a new category name in Name and/or select a symbolic color from the palette. 4. Click OK. Note that the category set of basic colors is not changed when its definition is modified. To add a category basic color 1. Select the required category from the Categories list. The current set of the selected category basic colors will appear in Color. 2. Click the button with eyedropper on Color and specify a point on the image. The indicated point color will be added to the Color list. If you make a mistake when basic color selecting, then delete it. The mistake in color selection is detected by examining the separation results that are shown in the preview window. To delete a category basic color 1. Select the required category from the list of Categories. 2. Click Delete in Color. To adjust separation sensibility to small objects Use the Noise Reduction slider and Despeckle option. Setting Despeckle on you can eliminate the influence of noise and small objects on the color analyzing procedures. The Noise Reduction slider allows redistribution of pixels by categories by modifying sensibility of separation to small objects. This parameter modifies the color averaging degree when calculating pixel colors. If averaging value is high, then small-sized droplet of unwanted colors (speckles) in large image objects will be classified as pixels of the categories to which relatively large objects belong. Due to high averaging value, the categories, containing small objects are distorted because the object edge dots are classified as objects of other categories. If averaging value is low, then the opposite effect is observed - the quality of small objects increases, the quantity of noise rises. By default the slider is set an average value. If categories that contain large filled areas are the most important to you, then reduce the value of this parameter (move to Low) to reduce raster speckles and holes in these category objects. If you try to get monochrome layers that contain the image of small or thin objects, such as texts, characters, level lines or grid, then increase the value of this parameter, to prevent the small object shapes from being thinned and distorted, caused by noise reduction. Of course, this increases the noise level on all layers. 106