KEYWORD border ptdist DESCRIPTION FindPoints Polygons FindPoints Polygons is similar to 2D Object Finder. It segments image data by assuming that pixels with zero intensity are not part of the features of interest and that the parts with positive intensity are. The edges of the features of interest are then traced and recorded as polygonal cross-sections. FindPoints Polygons has two additional capabilities: 1) FindPoints Polygons can work with the mask files generated by Find Points or Region Growing, which label separate regions with different values and use zero for points which are not in any region. Turn on the "Separate by mask" toggle when working with these files (unless you do want to lump together everything that is not background). 2) FindPoints Polygons will associate the polygonal cross-sections into 3D structures without the need to run the stand-alone application, Volume Builder. Volume Builder may still be useful for editing how the cross-sections are associated, taking measurements of the 3D structures, or to convert the 3D structures into a form that can be used in 3D Model or Clouds. Find Points and SurfHarmMod are two applications that can utilize the polygons and cross-section associations. FindPoints Polygons' user interface uses the same set of controls for selecting and processing a region as other Priism applications; consult Region.hlp for more information. There are additional controls to enter the name of the polygon (.Pol) file and the name of cross-section association (.Vol) file. The "exclude edge objs" toggle and "edge size" field control how objects which intersect the edges of an image are handled; these controls can be useful as a workaround for edge artifacts in the input images. The coarseness of the polygonal cross-sections, which has a strong effect on the time needed to manipulate the polygons in other programs, is set from the "dist between pts" field. If the input is to be treated as a binary mask (a positive value marks the pixels in the object of interest and a value of zero marks the pixels outside that object), then turn the "Separate by mask" toggle off. If the input has different regions marked by different positive values and you want to distinguish those regions in the output, turn the "Separate by mask" toggle on. FPPolygons accepts the command-line arguments described in Region.hlp. In addition, it has the following options: -binary Forces FPPolygons to treat the data as binary data: all positive values are grouped into the same object. -border=n Sets the number of pixels, n, along each edge that FPPolygons excludes from segmentation. n must be a positive integer. By default, FPPolygons excludes a one pixel wide border around the edge of the image. -exclude_edge If specified, FPPolygons filters out any polygons that have a vertex within n+1 pixels of the sides of the image where n is the border size set with -border. -poly=name Causes FPPolygons to generate a file named name to store the polygon results. If you do not use this option, FPPolygons does not save the polygons. -spacing=s Sets the distance between adjacent polygon vertices to be s pixels. By default, FPPolygons will generate polygon vertices spaced four pixels apart. -vol=name Causes FPPolygons to generate a file named name to store the information about how to link the polygons outlines to form volumes. If you do not use this option, FPPolygons does not save the connectivity information. The following example segments the contents of window 2 and saves the polygons in /var/tmp/2.pol and the object connectivity information in /var/tmp/2.bld: FPPolygons 2 -poly=/var/tmp/2.pol -vol=/var/tmp/2.bld border FPPolygons does not segment the points that are within n pixels of the sides of the image where n is the value in the "edge size" field. If the "exclude edge objs" toggle is on, FPPolygons filters out any polygons that have one or more vertices within n + 1 pixels of the sides of the image. ptdist Adjacent points in the polygonal cross-sections are separated by the distance, in pixels, from the "dist between pts" field. Using a a larger value gives coarser polygons but decreases the computational and drawing effort for programs that will use the polygons.