Incorrect VoxelSize in MetaMorph Reader #1966
Conversation
To fix the physical Z size, the difference between the absolute Z positions will need to be used. MetaMorph handler already parses the z positions. The physical z size is then calculated by subtracting the top and bottom values and dividing by the number of steps. Ticket details below: https://trac.openmicroscopy.org/ome/ticket/12977
| @@ -778,6 +778,32 @@ else if (differentZs) { | |||
| if (zDistances != null) { | |||
| stepSize = zDistances[0]; | |||
| } | |||
| else { | |||
| Vector<Double> zPositions = new Vector<Double>(); | |||
| Vector<Double> uniqueZ = new Vector<Double>(); | |||
There was a problem hiding this comment.
An ArrayList would suffice unless you need thread synchronization.
|
Note this PR is causing regressions on the Metamorph files of |
|
Note that if you configure the build parameters you can use https://ci.openmicroscopy.org/job/BIOFORMATS-5.1-merge-repository-subset/ (presumably someday DEV too) to do quick tests of these -- it does something similar to what is described on http://www.openmicroscopy.org/site/support/bio-formats5.1/developers/commit-testing.html. Often the problem is just a little bug in the code but one may sometimes legitimately need to generate adjustments for the |
Updated to use unique handlers in order to fix issues with merge tests https://ci.openmicroscopy.org/job/BIOFORMATS-5.1-merge-test_images_good/625/
The Physical Step Size is now only being calculated when each step in the z range is unique. If duplicate values appear within the range then we will not update the step size
|
Testing Steps:
nom1.nd
Expected physical z size for: |
Updating the physical z size calculations to use BigDecimal in order to control the precision of floating point arithmetic and remove rounding errors.
|
Tested using ImageJ. Everything works as expected. Please merge. |
Incorrect VoxelSize in MetaMorph Reader
To fix the physical Z size, the difference between the absolute Z
positions will need to be used.
MetaMorph handler already parses the z positions. The physical z size is
then calculated by subtracting the top and bottom values and dividing by
the number of steps.
Ticket details below:
https://trac.openmicroscopy.org/ome/ticket/12977
Splitting from #1964