IDR / idr-notebooks

Jupyter Notebooks for the Image Data Resource
https://idr.openmicroscopy.org/jupyter
BSD 2-Clause "Simplified" License
18 stars 19 forks source link

Stream images #123

Open ethancohen123 opened 2 years ago

ethancohen123 commented 2 years ago

Hi, Is it possible to read the images of a screen x through some stream methods (without any dowloading) ? For example, if I want to use two different screens without dowloading them but only read all images inside for some tasks, is it possible to do that ? Let me know if its not clear Thank you

will-moore commented 2 years ago

There's some examples at https://github.com/IDR/idr-notebooks/blob/master/IDR_API_example_script.ipynb for getting Images from a Screen and viewing Images. However, that will give you a "rendered" 8-bit rgb image, not the original pixel data.

To get the raw pixel values for an image, you'll need to use the Python OMERO api - see https://omero.readthedocs.io/en/v5.6.5/developers/Python.html#raw-data-access

You can get a connection conn to IDR as in the example at https://github.com/IDR/idr-notebooks/blob/master/Getting_Started.ipynb

Hope that helps,

Will

jburel commented 2 years ago

Another example analysing screen data in cellProfiler can also be found at https://github.com/ome/omero-guide-cellprofiler/blob/master/notebooks/ We use for that example data from idr0002

ethancohen123 commented 2 years ago

Thank you that is very relevant ! So if I understand correctly, it is totally possible to stream image data on a lab server (for example idr0002) and do analysis with them without any dowloading right ?

jburel commented 2 years ago

With the approach you will only stream the plane you wish to analyse. if you want to re-analyse you could also save the numpy array locally and inject them in your pipeline