<html><head><style type="text/css"><!-- DIV {margin:0px;} --></style></head><body><div style="font-family:times new roman, new york, times, serif;font-size:12pt"><div style="font-family: times new roman,new york,times,serif; font-size: 12pt;">Hi Luis,<br><br> Thank you for your reply. Please read on for answers to your questions.<br><div style="font-family: times new roman,new york,times,serif; font-size: 12pt;"><div><br>------------------><br><br>It shouldn't happen...<br><br><br> A bit more of information will be helpful...<br><br><br>For example:<br><br>1) Are the two synthetic images representing the same<br> Modality ?<br><br>-----------------> Yes, and intensity distribution is the same. The moving image is generated from the fixed image by rotation and translation.<br><br><br>2) What combination of registration components
did<br> you used for the Rigid registration ?<br><br> - What Metric ?<br> - What Transform ?<br> - What Interpolator ?<br> - What Optimizer ?<br><br>------------------> I use MeanSq metric, Versor3D Transform, Linear Interpolator and regular gradient descent optimizer.<br>The results after rigid registration are very good. Infact the final metric value reads 0.75. <br><br><br>3) Are you feeding DdeformableRegistration7 with the<br> resampled image resulting from a Rigid registration<br> process ? or did you set up the Bulk transform of<br> the BSpline deformable transform ?<br><br><br>------------------> Yes, I use the resampled image resulting from the rigid registration. The setBulkTransform was not available. I
see method in the .h file, but when I try to use it, VC++ gives me an error. VC++ could only list the private data member (mbulkTransform) but not the method to access it. So I use the resampled image. But since the moving image is derived from teh fixed, I dont see a problem here.<br><br>Also, when I compare the difference images from after performing rigid_registration to before_deformable_registration, the results are not the same. Do you think this is because of resampling as well? The metric value, obviously is very large (600+), but surprisingly the checkered images in one case were worse after registration.<br><br>To give you an example :<br> <br> One 3D registration took 13 hrs, 20 iterations and reduced the mean squares metric value from 533 to 22. Mean squares metric value after rigid registration was 0.75, the output of which was sent to deformableRegistration7.cpp<br><br>Here's the optimizer
output:<br>Starting Registration<br>Itr Value InfinityNormofProjectedGradient<br>0 533.946 47.9109<br>1 257.99 37.5374<br>2 146.285 19.8381<br>3 105.469 14.2091<br>4 78.1926 6.82423<br>5 64.971 5.15543<br>6 54.1814 3.72431<br>7 50.4374 5.19165<br>8 45.1611 2.137<br>9 41.2881 3.3166<br>10 36.9809 2.01088<br>11 34.4687 1.88032<br>12 31.1311 2.47016<br>13 29.6365 1.67621<br>14 27.6934 1.22647<br>15 26.9191 3.10447<br>16 25.5062 1.66605<br>17 24.6931 1.22317<br>18
23.9171 1.91149<br>19 23.4049 1.67905<br>20 22.1302 0.71955<br><br><br>4) How many iterations of the Deformable registration<br> did you manage to run before it stopped. ?<br><br>----------------->20<br><br>BTW, the size limit of emails to the list is 40Kb,<br> If you want to show images you could upload<br> them to a public web site and send the link<br> to the list.<br><br>--------------------> I'm sorry, I do not have access to a public site. If I can send you a tar file, I shall try to create one of < 40kb<br><br>Thank you for your time and patience,<br><br>Emma<br><br><br>Please let us know about the questions above.<br><br><br> Thanks<br><br><br><br> Luis<br><br><br>------------------<br>Emma Ryan wrote:<br>>
Hi,<br>> <br>> I am trying to register some synthetic data using <br>> deformableRegistration7.cxx from the ITK Examples. The synthetic data is <br>> almost well registered using rigid registration. So deformable <br>> registration is used to fix the slight mismatch. Unfortunately, the <br>> checkerboard output for the 'after' images are worse that 'before'.<br>> <br>> Why does this happen ?<br>> <br>> Any clues ? I haven't changed anything in the code except the Pixeltype<br>> <br>> Emma<br>> <br>> p.s. I tried to attach some images but it bounced back.<br>> <br>> ------------------------------------------------------------------------<br>> Need a vacation? Get great deals to amazing places <br>> <<a target="_blank"
href="http://us.rd.yahoo.com/evt=48256/*http://travel.yahoo.com/;_ylc=X3oDMTFhN2hucjlpBF9TAzk3NDA3NTg5BHBvcwM1BHNlYwNncm91cHMEc2xrA2VtYWlsLW5jbQ--%3Eon">http://us.rd.yahoo.com/evt=48256/*http://travel.yahoo.com/;_ylc=X3oDMTFhN2hucjlpBF9TAzk3NDA3NTg5BHBvcwM1BHNlYwNncm91cHMEc2xrA2VtYWlsLW5jbQ-->on</a> <br>> Yahoo! Travel.<br>> <br>> <br>> ------------------------------------------------------------------------<br>> <br>> _______________________________________________<br>> Insight-users mailing list<br>> Insight-users@itk.org<br>> <a target="_blank" href="http://www.itk.org/mailman/listinfo/insight-users">http://www.itk.org/mailman/listinfo/insight-users</a><br></div></div><br></div></div><br>
<hr size=1>Be a better Globetrotter. <a href="http://us.rd.yahoo.com/evt=48254/*http://answers.yahoo.com/dir/_ylc=X3oDMTI5MGx2aThyBF9TAzIxMTU1MDAzNTIEX3MDMzk2NTQ1MTAzBHNlYwNCQUJwaWxsYXJfTklfMzYwBHNsawNQcm9kdWN0X3F1ZXN0aW9uX3BhZ2U-?link=list&sid=396545469">Get better travel answers </a>from someone who knows.<br>Yahoo! Answers - Check it out.
</body></html>