Task #9191 (closed)
Opened 12 years ago
Closed 12 years ago
Bug: Deltavision channels wavelengths are not identified
| Reported by: | omero | Owned by: | mlinkert |
|---|---|---|---|
| Priority: | minor | Milestone: | Unscheduled |
| Component: | General | Version: | n.a. |
| Keywords: | n.a. | Cc: | carandraug+dev@…, l.mullee1@… |
| Resources: | n.a. | Referenced By: | n.a. |
| References: | n.a. | Remaining Time: | n.a. |
| Sprint: | n.a. |
Description
Hi
When opening a dv file (from a Deltavision microscope with softWoRx software) on ImageJ with the LOCI bioformats plugin and the "Colorize" option, the lookup table colors chosen are incorrect. Reading the documentation of the plugin I see that when the colors are not present on the metadata, the color mode defaults to "Composite" which I believe is what is happening (whatever is the the first channel gets colored red and the second green).
I have tested on softWoRx if the .dv.log file that pairs the .dv is needed for correct identification of the color. Removing the .dv.log file and opening the image on softWoRx, the colors are still picked correctly which suggests that the information is encoded on the actual TIFF/.dv file.
I have attached two images with red (mCherry filter) and green (GFP filter) taken on both orders (GFP-mCherry and mCherry-GFP order) to reproduce and for testing.
I have tested this on FIJI on Debian squeeze (LOCI plugins revision b4b47ad release 4.3-DEV).
The workaround in the lab is obviously to take the pictures in the red channel first and the green channel second. However, it took a long time before we understood what was going wrong since the 2 channels already don't have a big difference.
I would suggest to give a warning if the "colorize" option was chosen but had to default to "composite" due to the lack of metadata information to avoid catching users unaware.
Carnë
Attachments (4)
Change History (7)
Changed 12 years ago by omero
comment:1 Changed 12 years ago by jmoore
- Summary changed from Deltavision channels wavelengths are not identified to Bug: Deltavision channels wavelengths are not identified
- Type changed from Bug to Task
comment:2 Changed 12 years ago by mlinkert
- Component set to General
- Milestone set to Unscheduled
- Owner set to mlinkert-x
comment:3 Changed 12 years ago by mlinkert
- Resolution set to fixed
- Status changed from new to closed
The solution is to color the channels based upon the wavelength (and not assume that the first channel is red), just as softWorX and OMERO do.
Fixed with: https://github.com/melissalinkert/bioformats/commit/0ca42c7c65378163b851fe51d9845132d107bd1d
Note that new builds are not yet available.
red first, green second (dv file)