Laura Murgatroyd
Laura Murgatroyd
## Describe your changes Adds a bad pixel corrector which takes in a mask of bad pixel locations. On each projection, in every channel (if present) it performs a correction...
## Description This would be helpful as it allows you to adjust the windowing and ROI Currently if you run islicer with 2D data (i.e. a single reconstructed slice) you...
## Description E.g. we had set origin='bottom-left' and expected the origin to change. We should've instead been setting it to 'lower-left'. We should raise an error if the origin is...
## Describe your changes Modifies `Masker` so that the mean and median are calculated correctly along the axis, if specified. This is a quick fix to #1530 to fix the...
In an .xtekct file there is a section called 'Name' e.g.: ``` [XTekCT] Name=Proj ``` The NikonDataReader reads in this Name: https://github.com/TomographicImaging/CIL/blob/955ef6f2f5ca863e79c93e54dacf03f36fccfe52/Wrappers/Python/cil/io/NikonDataReader.py#L151 But it doesn't use this information at all....
Say I want to read a subset of a dataset to zoom in on an area of a sinogram: ```python grey_data = NikonDataReader(file_name = GREY_DATA_NIKON_FILE, roi={'angle': (930, 980)}).read() show2D(grey_data) ```...
Docs for AcquisitionGeometry: https://tomographicimaging.github.io/CIL/nightly/framework.html?highlight=ndim#acquisitiongeometry It would be useful to list the properties of AcquisitionGeometry in the docs. Methods are documented here for configuring and using the geometry: https://tomographicimaging.github.io/CIL/nightly/framework.html?highlight=ndim#configure-the-geometry However, the...
Looking at this documentation I noticed that we use 'center' rather than 'centre' in the Function methods: https://github.com/TomographicImaging/CIL/blob/ff4741092b7f10bde5e0572c0ac103288a163ffc/Wrappers/Python/cil/optimisation/functions/Function.py#L151 This is inconsistent with our: https://github.com/TomographicImaging/CIL/blob/ff4741092b7f10bde5e0572c0ac103288a163ffc/Wrappers/Python/cil/processors/CentreOfRotationCorrector.py#L25 And we should be consistent across...
Add a processor to automatically run a bad pixel correction on projections. This will take the mean of nearest neighbours (including diagonals)
## Description I created an environment using the instructions in the readme (on a linux machine) > conda create --name cil -c conda-forge -c https://software.repos.intel.com/python/conda -c ccpi cil=24.1.0 astra-toolbox=*=cuda* tigre...