I am trying to detect mitochondria. I use simple treshold to produce binary image and then I select primary objects with manual treshold 0.5. It works in version 2.2(ubuntu 16 LTS), in version 3 (ubunty 14 LTS) I get 0 objects. All the detected objects have yellow bondary - I guess it is because the max number of objects is exceeded, it is not problem with touching of the boundaries. In CP2.2 I get thousands. Could you give me an advice what to do?
I guess but I am not sure that I already faced similar problem with with CP2.2, so acceptance of higher number of objects by CP2 could be a result of some tinkering. But I cannot find any tips what should be changed. thank you
What commit/tag are you using in 3.0? It sounds similar to a bug that has since been fixed but I can’t say whether or not that’s your issue without more information.
I installed CP according to instructions here
its 3.0.0.rc1. thanks
Hi, can you try updating to the v3.0.0 tag and see if that works? I’m pretty sure that’s the old bug I was thinking of, which has been fixed since rc1
I installed newer ubuntu. If I use the instruction here https://github.com/CellProfiler/CellProfiler/wiki/Source-installation-(Ubuntu-16.04-LTS)
should I use the “git checkout stable” line?
I do not understand those github branchings - if a bux is fixed, is the fix propagated to the stable version? If not how can I choose ther right version from command line?
We’ll probably update the “stable” to v3.1.4 when we release that (hopefully very soon!), but for now the stable is still v 2.2.1 I think. If you use
git checkout v3.0.0 you’ll get the branch I meant.