Hi all,
I am using tetgen to construct a tetrahedral mesh with the extracted pial surface in subject 004 from the tutorial datasets. But I got an error message from tetgen to say that some triangles are intersecting each other. I am not sure whether anyone has found this before. The file I tested was 004/lh.pial. I first transferred it to asc format with mris_convert, and then make a straightforward transfer to .off (used by tetgen and geomview). I loaded the .off file with geomview and it looks right. Then I run tetgen with the .off file and got an error message like below, which means that there are intersections among the facets. Error: Invalid PLC. Two subfaces (219028, 5283, 5282) and (219054, 5300, 5283) are found intersecting each other. Hint: Use -d switch to find all intersecting facets.
This is a little strange to me since the topology issue has been handled in the reconstructed surface, does any one have any clue?
Thank you very much!
Yasheng
Hi Yasheng,
hmmm, there is triangle/triangle intersection code that is supposed to prevent this from happening. Does this happen if you use the native (binary) format? How many faces intersect?
Bruce
On Mon, 8 Mar 2010 yasheng_chen@med.unc.edu wrote:
Hi all,
I am using tetgen to construct a tetrahedral mesh with the extracted pial
surface in subject 004 from the tutorial datasets. But I got an error message from tetgen to say that some triangles are intersecting each other. I am not sure whether anyone has found this before. The file I tested was 004/lh.pial. I first transferred it to asc format with mris_convert, and then make a straightforward transfer to .off (used by tetgen and geomview). I loaded the .off file with geomview and it looks right. Then I run tetgen with the .off file and got an error message like below, which means that there are intersections among the facets.
Error: Invalid PLC. Two subfaces (219028, 5283, 5282) and (219054, 5300, 5283) are found intersecting each other. Hint: Use -d switch to find all intersecting facets.
This is a little strange to me since the topology issue has been handled in the reconstructed surface, does any one have any clue?
Thank you very much!
Yasheng
freesurfer@nmr.mgh.harvard.edu