Mantid Imaging 2.2

This release brings some new features and bug fixes.

The key new features of 2.2 are NeXuS import and a new reconstruction algorithm, Primal-Dual Hybrid Gradient from the Core Imaging Library (CIL).

New features

  • #927 : ROI norm: use averaged view in ROI selector

  • #925 : Jenks / Otsu everywhere

  • #926 : Manual arithmetic filter

  • #988 : Use averaged view in ROI selector for crop

  • #964 : Open NeXus files

  • #1010 : Write slope and offset into int tiffs

  • #987 : Allow Flat-fielding without dark subtraction

  • #1011 : Pixel size should allow setting decimal places

  • #1013 : NeXus Loading Window

  • #904 : Default to using image center for ring removal COR

  • #1025 : Add the Total Variation (TV) with Primal-Dual Hybrid Gradient (PDHG) from CIL

  • #1027 : Flat field warning if output has negative values

  • #1027 : NeXus Loader: Message when loading large files

  • #1026 : Recon warn if input has NaNs

  • #816 : NaN Removal Filter

  • #1035 : NeXus Loader: Select every Nth image

  • #1035 : List of bad slices in error message could be compressed

Fixes

  • #939 : median: NameError: name ‘cp’ is not defined

  • #957 : Crop Coordinates ROI not changing

  • #961 : Test failure: ImportError: libcuda.so.1

  • #959 : Median GPU does not work with even values

  • #990 : Remove combine histogram checkbox

  • #989 : Remove legend checkbox

  • #1001 : Exception when selecting Log in Load Dataset dialog

  • #991 : Handle longer running previews better

  • #1021 : Prevent simultaneous Astra calls

  • #1036 : Exception in ReconstructWindowView.show_recon_volume prevents recon closing

  • #1046 : Can’t rotate NeXus images

  • #1048 : NeXus Loader: ValueError: Illegal slicing argument for scalar dataspace

  • #1055: Arithmetic operation breaks flat-fielding

  • #898: Improve user documentation for operations

  • #1059: Crash in recon Minimise Error with CIL

  • #1032: Crash in RingRemovalFilter

  • #1057: Fix crash when refining COR

  • #1056: Cannot set range [nan, nan] in ROI normalisation

  • #1019: ERROR: Presenter error: Error applying filter for preview - when closing stacks

  • #1068: Don’t show misleading success message when operation cancelled

  • #1082: CIL show memory warning

  • #1014: KeyError when closing stacks

  • #1029: Segmentation fault in image_in_vb

  • #1081: Error if pressing preview button before selecting a dataset

  • #940: Ops window, pixel value not updating for part of image

  • #1093: TIFF output blank if NaN in stack

  • #1095: AttributeError loading 180 deg projection

  • #1110: Segmentation fault COR table refine

  • #1115: Copy everything from metadata dialog

  • #1112: Minimise Error, number of slices input should have minimum of 2

  • #1067: Failed to load stack “log file is not recognised”

  • #1121: FF when negative values in output, no completed message

Developer Changes

  • #917 : Intermittent failure of StripeRemovalTest.test_memory_executed_wf

  • #980 : Use pre-commit

  • #978 : Update Qt API usage

  • #998 : setup.py command to install current developer requirements

  • #969 : Improve mypy coverage in core.io

  • #1007 : Change applitools check level to content

  • #1086 : GUI system tests

Dependency updates

  • pyqtgraph 0.12, scikit-image 0.18, tomopy 1.9, numpy 1.18