Hi Mark
have you tried "conforming" the inputs (making them all 256^3, 1mm iso and 8 bits/voxel)? Bruce On Fri, 11 Jul 2014, Mark Plantz wrote:
Hello freesurfers, I am running into some sort of memory allocation issue when running mri_ca_train. I am simply using an intensity volume and a segmentation volume to create a GCA atlas. I keep getting this "segmentation fault: 11", which is data allocation error. I have tried using all different environments and can't avoid it. The output is below:
sh-3.2# mri_ca_train -T1 withSkull.mgz -seg seg_edited.mgz atlas_test /hearingbrain/Desktop/atlas_withSkull.gca reading T1 data from subject's mri/withSkull.mgz directory reading segmentation from subject's mri/seg_edited.mgz directory training on 1 subject and writing results to /Volumes/Macintosh HD/Users/hearingbrain/Desktop/atlas_withSkull.gca
processing subject atlas_test, 1 of 1... reading input 0: /Applications/freesurfer/subjects/atlas_test/mri/withSkull.mgz gca width modified from 256 to 181 gca height modified from 256 to 217 gca depth modified from 256 to 181 Segmentation fault: 11
Any ideas?
-MP