bmiadmin@mitchelllab-linux:~/Tamara/FreeSurfer_GENFI/analysis/Register$ mri_robust_register --mov rawavg_1.mgz --dst rawavg_2.mgz --lta v1to2.lta --affine --satit
$Id: mri_robust_register.cpp,v 1.52 2011/03/16 21:23:49 nicks Exp $

--mov: Using rawavg_1.mgz as movable/source volume.
--dst: Using rawavg_2.mgz as target volume.
--lta: Output transform as v1to2.lta . 
--affine: Enableing affine transform!
--satit: Will iterate with different SAT to ensure outliers below wlimit!

reading source 'rawavg_1.mgz'...
reading target 'rawavg_2.mgz'...

Registration::setSourceAndTarget ...
   Mov: (1, 1, 1.1)mm  and dim (256, 256, 208)
   Dst: (1, 1, 1.1)mm  and dim (256, 256, 208)
   Asserting both images: 1mm isotropic and (256, 256, 229) voxels
   Original : (1, 1, 1.1) mm size and (256, 256, 208) voxels.
   Resampled: (1, 1, 1) mm size and (256, 256, 229) voxels.
   Reslicing using trilinear interpolation 
   Original : (1, 1, 1.1) mm size and (256, 256, 208) voxels.
   Resampled: (1, 1, 1) mm size and (256, 256, 229) voxels.
   Reslicing using trilinear interpolation 


 Registration::findSaturation 
   - Gaussian Pyramid 
   - Gaussian Pyramid 

   - Max Resolution used: 2
     -- gpS ( 64 , 64 , 57 )
     -- gpT ( 64 , 64 , 57 )
   - computing centroids 
   - computing initial transform
     -- using translation info
   - running loop to estimate saturation parameter:
     -- Iteration: 1  trying sat: 16
         min sat: 16 ( 0.349663 ), max sat: 0 ( -1 ), sat diff: -16
     -- Iteration: 2  trying sat: 32
         min sat: 32 ( 0.18235 ), max sat: 0 ( -1 ), sat diff: -32
     -- Iteration: 3  trying sat: 64
         min sat: 32 ( 0.18235 ), max sat: 64 ( 0.060566 ), sat diff: 32
     -- Iteration: 4  trying sat: 48
         min sat: 32 ( 0.18235 ), max sat: 48 ( 0.100221 ), sat diff: 16
     -- Iteration: 5  trying sat: 40
         min sat: 32 ( 0.18235 ), max sat: 40 ( 0.13302 ), sat diff: 8
     -- Iteration: 6  trying sat: 36
         min sat: 32 ( 0.18235 ), max sat: 36 ( 0.155648 ), sat diff: 4
     -- Iteration: 7  trying sat: 34
         min sat: 34 ( 0.168781 ), max sat: 36 ( 0.155648 ), sat diff: 2
     -- Iteration: 8  trying sat: 35
         min sat: 35 ( 0.162319 ), max sat: 36 ( 0.155648 ), sat diff: 1
     -- Iteration: 9  trying sat: 35.5
   - final SAT: 35.5 ( it: 9 , weight check 0.158663 <= 0.16 )


 Registration::computeMultiresRegistration 
   - computing centroids 
   - computing initial transform
     -- using translation info
   - initial transform:
Ti = [ ...
 1.0000000000000                0                0  7.8421342857496 
               0  1.0000000000000                0 -23.5616252396979 
               0                0  1.0000000000000  1.2809103259631 
               0                0                0  1.0000000000000  ]

   - initial iscale:  Ii =1

Resolution: 3  S( 32 32 28 )  T( 32 32 28 )
 Iteration(f): 1     -- diff. to prev. transform: 3.30263
 Iteration(f): 2     -- diff. to prev. transform: 0.531007
 Iteration(f): 3     -- diff. to prev. transform: 0.170753
 Iteration(f): 4     -- diff. to prev. transform: 0.0964301
 Iteration(f): 5     -- diff. to prev. transform: 0.11475 max it: 5 reached!

Resolution: 2  S( 64 64 57 )  T( 64 64 57 )
 Iteration(f): 1     -- diff. to prev. transform: 3.45373
 Iteration(f): 2     -- diff. to prev. transform: 0.310855
 Iteration(f): 3     -- diff. to prev. transform: 0.0868077
 Iteration(f): 4     -- diff. to prev. transform: 0.0409148
 Iteration(f): 5     -- diff. to prev. transform: 0.0138362 max it: 5 reached!

Resolution: 1  S( 128 128 114 )  T( 128 128 114 )
 Iteration(f): 1     -- diff. to prev. transform: 2.07652
 Iteration(f): 2     -- diff. to prev. transform: 1.18617
 Iteration(f): 3     -- diff. to prev. transform: 0.0244597
 Iteration(f): 4     -- diff. to prev. transform: 0.0036059  < 0.01  :-)

Resolution: 0  S( 256 256 229 )  T( 256 256 229 )
 Iteration(f): 1     -- diff. to prev. transform: 0.455513
 Iteration(f): 2     -- diff. to prev. transform: 0.0318535
 Iteration(f): 3     -- diff. to prev. transform: 0.00858  < 0.01  :-)

   - final transform: 
Tf = [ ...
 1.0109407600195 -0.0009267109018  0.0053997282799  1.2184107520865 
-0.0092283795014  1.0113299213890 -0.0135809690517 -14.8648753544432 
-0.0044243147587  0.0117414476923  1.0107664670226 -1.7224801501516 
               0                0                0  1.0000000000000  ]

   - final iscale:  If = 1

Final Transform:
Adjusting final transform due to non isotropic voxels ...
M = [ ...
 1.0109407001227 -0.0009267399430  0.0059396475320  1.2189587985683 
-0.0092283667133  1.0113299188150 -0.0149390077859 -14.8662677851001 
-0.0040220715641  0.0106743077812  1.0107664666105 -1.5649209924601 
               0                0                0  1.0000000000000  ]

 Determinant : 1.03358

 Decompose into Rot * Shear * Scale : 

Rot = [ ...
 0.9999795758934  0.0041348506255  0.0048734798989 
-0.0040728637478  0.9999115484027 -0.0126612460613 
-0.0049254011930  0.0126411384470  0.9999079667859  ]

Shear = [ ...
 1.0000000000000 -0.0050409813618  0.0010109298245 
-0.0050429465396  1.0000000000000 -0.0021128755499 
 0.0010108439155 -0.0021118727042  1.0000000000000  ]

Scale = diag([  1.0109774487584  1.0113715685396  1.0108915357302  ])


writing output transformation to v1to2.lta ...
converting VOX to RAS and saving RAS2RAS...

To check transform, run:
  tkregister2 --mov rawavg_1.mgz --targ rawavg_2.mgz --lta v1to2.lta --reg v1to2.reg

Registration took 3 minutes and 57 seconds.