Repository navigation
mtex-7.1.0
MTEX 7.1 brings three-dimensional EBSD, the alignment of EBSD maps with electron images, an HDF5 export that keeps the vendor's file, and a long list of corrections found while rewriting the documentation. The changelog has the full list; the technical changelog the causes and numbers behind each fix.
Three-Dimensional EBSD
A measured volume from DREAM.3D or Xnovo GrainMapper3D loads into an EBSD3 object and is analysed with the commands of the planar case:
ebsd = EBSD3.load('SmallIN100_MeshStats.dream3d')
[grains,ebsd] = calcGrains(ebsd,'angle',5*degree)
grains = smoothBoundary(reduceBoundary(grains,2,'quadric'),taubinFilter(20))calcGrains reconstructs three-dimensional grains from the voxels and returns a closed surface mesh with the misorientation on every face. reduceBoundary and smoothBoundary coarsen and smooth the voxel staircase with the planar boundary filters, keeping triple lines and quadruple points in place; edges and nodeType expose the boundary network; slice cuts any plane out of a volume as an ordinary EBSD map. Chapter: Three-Dimensional EBSD Analysis.
TrueEBSD
trueEbsd aligns a sequence of images of the same area, an EBSD map and its electron images, onto one grid, correcting the tilt and drift of the scan. Built from mapImage, spatialTransform, xcfShift and a gridLayout for gridify. Method by Tong et al., arXiv 2605.00703. Page: TrueEBSD.
EBSD Export That Keeps the File
export(ebsd,'file.h5oina') copies the file the data was imported from and writes only the changed data into the copy, so patterns and acquisition settings pass through untouched. The .ang and .ctf exporters reproduce the angles of the original file on a round trip.
Also New
calcTaylorreturns the plastic work, andcalcLankfordminimises it- the documentation has been rewritten page by page into a textbook of texture analysis that uses MTEX
doubleof an orientation embedding takes its coordinates from an orthonormal basis computed from the symmetry group
Corrections Worth Knowing
- the Radon transform of a fibre ODF under a nontrivial specimen symmetry averaged over the crystal orbit only
equispacedSO3Griddelivered a quarter of the points asked for under 222 or mmm specimen symmetry- the weights of a symmetrised
SO3FunRBFfollowed the wrong centers - the angle and logarithm of a very small rotation floored at about 2e-6 degree
- the Siemens D5000 and Seifert
.njapole figure readers merged or dropped measurements - grain boundary plots came out garbled on R2024b and later, most visibly on R2026a
sliceof a volume returned the plane through the origin whatever plane it was given
Compatibility
The import wizard needs MATLAB R2022b or newer; the command-line import works on older releases as before.
Contributions
- Remove duplicate "EBSDOrientationPlots" from TOC by @Tijmenvermeij in #2594
- update mtex_settings.m for deployed app by @vtvivian in #2497
- Dockerized mtex by @martinherrerias in #2494
Full Changelog: mtex-7.0.0...mtex-7.1.0