I am sorry, but I do not understand how to load several images into the MakeProjection module. If I have a stack it works well, but not if I have individual images. My problem is that my stack is of an organ of containing many objects and I wish to cut the stacks into sub stacks and then have a max projection done. I have unpacked the stack into its constituent images, and I load them into a new pipeline, but there I cannot assemble several images into one projection. Any suggestions?
I’m not sure I quite understand where the problem lies from your description. Once you load the images and name them in LoadImages (call them ‘Image’ for example, if you’re using one channel of data). Then ‘Image’ is used as input into MakeProjection, with a named result (say, ‘Projection’); the projection is completed at the end of the analysis run. If you wanted to save it, you would use SaveImages with Projection as input and select “Last cycle” under “Select how often to save”. Is this something different than what you are doing?
you have understood my problem, except what you describe does not work. When I load the images, for instance image 1 and image2, the Makeprojection only allows me to to choose either image 1 or image2. If I load both images and name them both image, no projection is done either. The module seems to work only with stacks, but I would like to subdivide a large stack in several projections. Hope this is clearer, else I can upload two files with an example pipeline.
In LoadImages, it sounds like you are loading the multiple images into more than one channel name in CP. You should use only one channel name for each projection. E.g. if your image filenames to make a projection are Image1.jpg and Image2.jpg, then in LoadImages use Exact filenaming and simply a single channel name ‘Image’, so that it captures all the images to be projected into one “channel”. Then in MakeProjection you simply choose ‘Image’ and the projection should work.
If I am misunderstanding, please upload your pipeline and a couple example images.
I think I’m doing exactly what you are telling me, but the image I receive is always the same. I uploaded an example pipeline and the three images I tried it on (at the bottom of the message). Irrespective of the operation I do, I receive at the end just the third and last image loaded. Additionally I get an error message that I just ignore when loading with the same name:
Traceback (most recent call last):
File “/cc/release_10953/CellProfiler/cellprofiler/pipeline.py”, line 309, in run
File “/cc/release_10953/CellProfiler/cellprofiler/modules/loadimages.py”, line 2083, in run
File “/cc/release_10953/CellProfiler/cellprofiler/measurements.py”, line 315, in add_measurement
AssertionError: Feature Image.MD5Digest_DNA has already been set for this image cycle
Any help is welcome.
I was wondering whether you had time to look at the images and pipeline. I would need to do this projection for a project.
I have attached a pipeline which works. There were a couple fundamental issues that you can now see in the attached pipeline:
(1) LoadImages works much more simply (albeit intelligently!) than you had configured. You simply need a single channel (e.g. “DNA” – never replicate the same name) and then to choose, as the setting says, “Text that these images have in common”. E.g. I chose “2011-06-21” for your images, however you may need to be more specific in your full experiments.
(2) The 3 SaveImages were all set to save the same image name, “ProjectionBlue”. I changed the second and third to ProjectionBlue2 and ProjectionBlue3, resp.
The outputs look reasonable to me, especially the Max. The Power and Variance do look the same in a quick visual inspection, but we do say that the “Power” method is ‘experimental’ Note also that the latter two are very dim, in an absolute sense, and that you need to use some sort of scaling to see anything (as CP does by default, but other viewers might not).
Hope this helps,
Make_Projection_Fix.cp (4.84 KB)
OK, I see I misunderstood how the node works. Thank you very much, it will help me in my analysis.
Glad to hear it. Do let us know if we can clarify anything else.