From: mailbox@dmitry-kazakov.de   
      
   On Thu, 22 Sep 2005 11:24:25 GMT, makc.the.great@gmail.com wrote:   
      
   > I am now trying to put together some code to vectorize scanned maps.   
   > There is basically one issue that puzzles me, line detection.   
   > I have read numerous http articles on filter-based detectors, but they   
   > all seem to say YES/NO to a question, if there is a line. I need   
   > (x,y)-(x,y) response. Other articles deal with photoimages; they are   
   > focused on finding at least some lines anywhere. In my case I think I   
   > have to use an information I have as to what lines look like, while   
   > those mentioned approaches do not use such an information (primarily   
   > because they don't have it). So, after days, I feel like in empty   
   > dark room. Would someone please turn lights on for me?   
      
   Here is a somewhat complementary approach to one described by Ted Dunning:   
   run segmentation through the image. For example, using a region growing or   
   any other method. For scanned maps almost any segmentation method should   
   work well. If the number of colors is limited, you can even do clustering   
   based on the histogram.   
      
   Anyway, the borders between the segments are your lines. Of course, small   
   segments should be merged with the big ones or removed. Now the end points   
   would be ones where several segments meet.   
      
   --   
   Regards,   
   Dmitry A. Kazakov   
   http://www.dmitry-kazakov.de   
      
   [ comp.ai is moderated.  To submit, just post and be patient, or if ]   
   [ that fails mail your article to , and ]   
   [ ask your news administrator to fix the problems with your system. ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)