software_issues
software_issues > RE: Problems using TetraMetrix?
Jan 2, 2012 09:01 PM | Carl Lederman
RE: Problems using TetraMetrix?
Hi everyone. There are two issues that people have contacted me
about over the last few months.
1. The data types for the .img and .hdr files that are supported are: SIGNED SHORT, SIGNED INT AND SIGNED FLOAT. (so for example, double is not supported). So before using this software you will simply need to convert the image to one of the supported types.
2. If the size of the image is too big, the computer you are using may run out of memory. ( For reference, about 10 million voxels worked OK with 8 GB of RAM). If the objects within the image, ie the non-zero voxels, do not push right up against all edges of the image, one simple solution would be to crop off some of the 0-labeled voxels. After the mesh is created the nodes could simply be moved in the x, y, or z directions to match the corrodinates of the mesh with the original image.
So, unfortantely I am working full time in another area right now and I may not be able to make a comprehensive update to the code. But, thankfully, a number of people have been able to successful run the existing 'beta' code by following the one page instructions on this site. And I am still willing to make some future modifications based on the level of need and interest.
Carl
1. The data types for the .img and .hdr files that are supported are: SIGNED SHORT, SIGNED INT AND SIGNED FLOAT. (so for example, double is not supported). So before using this software you will simply need to convert the image to one of the supported types.
2. If the size of the image is too big, the computer you are using may run out of memory. ( For reference, about 10 million voxels worked OK with 8 GB of RAM). If the objects within the image, ie the non-zero voxels, do not push right up against all edges of the image, one simple solution would be to crop off some of the 0-labeled voxels. After the mesh is created the nodes could simply be moved in the x, y, or z directions to match the corrodinates of the mesh with the original image.
So, unfortantely I am working full time in another area right now and I may not be able to make a comprehensive update to the code. But, thankfully, a number of people have been able to successful run the existing 'beta' code by following the one page instructions on this site. And I am still willing to make some future modifications based on the level of need and interest.
Carl
Threaded View
Title | Author | Date |
---|---|---|
Carl Lederman | Apr 21, 2011 | |
Carl Lederman | Jan 2, 2012 | |