@AddEdge not running correctly

AFNI Message Board
1

Hi there,

I am using afni_proc.py for some pre-processing. I added -AddEdge as an option as part of my align_epi_anat block.

This has successfully generated a folder of edge images, but when it has run and I cd into the AddEdge folder, when I try to run @AddEdge -auto it throws up a bunch of different overlaid images one after the other, all disappearing quickly, leaving only my original anatomical image as the background with cyan edges, and a big block cyan blob (not aligned correctly) as my epi. My epi is a partial coverage slab, which is why they aren’t aligned, but not sure what else is going on for this review script to be acting strangely.

When I view the files separately by opening afni on it’s own and setting the overlay and underlay, this works ok, but I can’t work out how to colour the underlay, overlay and overlap differently so I can actually see how well aligned the images are (currently are greyscale).

Any help would be appreciated, thank you!

Harriet

code (in case it is useful)

afni_proc.py -subj_id $subj
-blocks align
-dsets $data_dir/P007_FR_RUN*
-tcat_remove_first_trs 0
-copy_anat $data_dir/FR_uniden_avg
-anat_has_skull yes
-align_opts_aea
-deoblique on
-ginormous_move
-partial_coverage
-AddEdge
-check_flip

2

Sorry about that. There was a change in the way the AFNI GUI does its scaling, and this script wasn’t updated to match. I’ve put in a couple small fixes for this. They will be available in the next binary distribution. In the meantime, you can toggle the autorange on and off in the overlay panel to see the images. Also review the QC directory for the images there that show alignment. Also @snapshot_volreg and @chauffeur_afni can generate similar edge enhanced images.

3

… and just to note, we built the binaries last night, so this fixed version should be available NOW NOW NOW NOW NOW!

Get 'em while they are fresh and hot out of the oven.

–pt