XPost: sci.image.processing, comp.graphics.algorithms, comp.ai.neural-nets   
   From: pangfu@hotmail.com   
      
   "TathagataC"  wrote in message   
   news:bg5nhp$3d7$1@mulga.cs.mu.OZ.AU...   
   > >   
   > > What do people think? How expensive is (2) ? Is it overkill? What kind   
   > > of signatures would be possible for (3)?   
   > >   
   >   
   > use the 2d discrete fourier transform to compute the signatures. use   
   > only the phase for comparison if necessary (to neglect variations in   
   > intensity). this should be pretty fast to implement and test. we have   
   > tried it on 64x64 gray scale images and got good results.   
   >   
   > regards   
   > tathagatac   
   >   
      
   Tathagatac,   
      
   Do you have any reference (paper, code,samples etc) for your 2d FFT   
   signature based classification problem/solution?   
      
   Cheers..   
   nospam_pangfu@hotmail.com   
      
   [ 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)