kikuchipy: an open-source toolbox for analysis of EBSD patterns

Håkon W. Ånes, Phillip Crout, Lars Andreas Lervik, Ole Natlandsmyr, Tina Bergh, Jarle Hjelen, Antonius T. J. van Helvoort, Knut Marthinsen · 2026

Read kikuchipy: an open-source toolbox for analysis of EBSD patterns by Håkon W. Ånes, Phillip Crout, Lars Andreas Lervik, Ole Natlandsmyr, Tina Bergh, Jarle Hjelen, Antonius T. J. van Helvoort, Knut Marthinsen (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-25722.txt.

Introduction

Ånes and colleagues present kikuchipy, an open Python toolbox for electron backscatter diffraction patterns. It handles Hough and dictionary indexing, plus orientation and projection-center refinement, for all major vendor formats. Results can be checked with index-independent maps and by matching experimental with simulated patterns. The team demonstrates it on super duplex stainless steel, on Al-Si phases in a cast alloy, and on particle phases in an Al-Mn alloy. Patterns and workflows are public, and the tool is built as a community resource anyone can extend.

Catalog record

CREATORHåkon W. Ånes, Phillip Crout, Lars Andreas Lervik, Ole Natlandsmyr, Tina Bergh, Jarle Hjelen, Antonius T. J. van Helvoort, Knut Marthinsen
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE68574
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2605.25722

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

kikuchipy: an open-source toolbox for analysis of EBSD patterns Håkon W. Ånes, Phillip Crout, Lars Andreas Lervik, Ole Natlandsmyr, Tina Bergh, Jarle Hjelen, Antonius T. J. van Helvoort, Knut Marthinsen Abstract We present kikuchipy, an open-source toolbox for analysis of electron backscatter diffraction patterns, written in Python. The software is capable of both Hough and dictionary indexing and orientation and/or projection center refinement of patterns stored in file formats from all major vendors. Indexing results can be validated using maps independent of indexing and by visually comparing experimental and simulated patterns. By leveraging scientific packages in the Python ecosystem, emphasis is put on making the indexing workflow flexible and improve results through fast iteration. The software’s capabilities are demonstrated on three application examples: analysis of orientation relationships in a super duplex stainless steel, phase differentiation of aluminium and silicon in a cast modified Al–Si alloy, and phase differentiation of particles in an Al–Mn alloy as Al 6 Mn or α \alpha –AlMnSi. The diffraction patterns and analysis workflows are made publicly available. kikuchipy was created and is developed as a resource for the electron microscopy community, allowing anyone to improve the software or include it into their own analysis workflows or softwares. Introduction Electron backscatter diffraction (EBSD) is a technique to characterize crystallographic features in the scanning electron microscope (SEM) 67 . A typical setup for acquisition of an EBSD dataset within the SEM is illustrated in Fig. 1 . An electron beam is rastered in a grid across a sample surface while collecting a two-dimensional (2D) electron backscatter pattern (EBSP) per beam position. EBSD is most often used to estimate the orientation of the strongest scattering crystal within the interaction volume in each beam position. This is called indexing and results in rich surface maps. Based on pioneering work by research groups at Yale 80 and Risø 35 , indexing is automated and widely available, typically in commercial solutions. The active development of indexing with openly available algorithms has been limited, although a few open-source projects exist, notably the EMsoft suite of command-line programs 25 , the EMSphinx graphical user interface (GUI) 39 , the AstroEBSD MATLAB software 11 , and the PyEBSDIndex Python software 64 . With kikuchipy, we present a widely available and accessible Python toolbox for further development of EBSP analysis. Figure 1: EBSD acquisition within the SEM. There are many sources of error when indexing an EBSP 60 ; 59 . Incorrect indexing may occur unexpectedly, and it might be hard to identify why. Examples of unexpected results are mis-indexing different crystallographic phases as the same phase, or a scatter of single points of highly different orientations within one grain or along a grain boundary. It is crucial that the indexing steps are transparent and that we are able to apply them in the desired order when troubleshooting problematic cases. Complete insight is generally unavailable for commercial solutions but is available for the mentioned open-source software. However, it is our experience that the software forsake flexibility to increase user-friendliness in terms of automation and ease-of-use. kikuchipy addresses this issue. The software’s primary aim is to make the off-line indexing workflow—from inspecting the acquired patterns to validating the indexing results—flexible and iterable. Analysis workflows can be built with flexibility since functionality is provided as independent tools. Indexing steps can then be optimized to identify and eliminate sources of error. Hough indexing and pattern matching approaches 15 ; 51 ; 73 are available. kikuchipy has already been used to improve the signal-to-noise ratio (SNR) of EBSPs 3 and to characterize a range of materials: steels 1 ; 12 , a nickel-based superalloy 2 , aluminium (Al) alloyed with manganese (Mn) 4 ; 5 , an Al–steel weld 8 , and ceramics 66 ; 65 . As with previous open-source EBSD software 56 , kikuchipy is developed as a research and development tool for the EBSD community. As such, emphasis is put on flexibility rather than high indexing speeds. This paper showcases the capabilities of kikuchipy. The focus is on the workflow of indexing an acquired EBSD dataset. Fig. 2 depicts a general analysis workflow using kikuchipy. The workflow consists of initial steps of loading and inspecting patterns, indexing steps to be optimized, and steps of visual validation and final saving of the indexing results. We give three examples of applying this workflow to challenging metallurgical datasets using kikuchipy. Example I demonstrates how to characterize and visualize orientation relationships in a super duplex stainless steel (SDSS) with a substantial fraction of undesirable secondary phases. Example II addresses the issue of differentiating between similar crystallographic phases using EBSD only by indexing aluminium and silicon in an Al–Si alloy. Example III concerns phase differentiation of micron-sized second-phase particles in an Al–Mn alloy as Al 6 Mn or \alpha –AlMnSi. The raw EBSD data and workflow notebooks are made publicly available for verification and reuse from Zenodo at https://doi.org/10.5281/zenodo.20290635 . Figure 2: Typical workflow for off-line indexing of a dataset of EBSPs using kikuchipy. Software capabilities and workflow kikuchipy is implemented in Python and supports Windows, macOS, and Linux. The software can be installed from the Python package index (PyPI) manager, the Anaconda package manager, or from source at https://github.com/pyxem/kikuchipy. kikuchipy extends the HyperSpy library 19 ; 20 for EBSP analysis. HyperSpy is established within the electron microscopy community and provides tools for interactive analysis of multi-dimensional datasets. kikuchipy follows HyperSpy’s object-oriented design, with most analyses available as methods of an EBSD class. Many file readers are available, enabling file format-agnostic workflows that can be shared with colleagues or alongside a publication. Datasets of sizes larger than available computer memory can be analyzed in parallel using Dask 63 . To reduce computation time, performance-critical parts are compiled with Numba’s just-in-time compiler 34 . Two other libraries are essential to kikuchipy’s functionality: orix 28 ; 6 for handling rotations and vectors using crystal symmetry and diffsims 27 for parts of the diffraction simulations. Users interact with kikuchipy by interactive scripting in a notebook 31 or console or by running a script. The documentation 1 1 1 https://kikuchipy.org . provides many examples and tutorials for users to build upon when creating workflows suited to their analyses. The documentation includes a complete description of all functionality and parameters as well. Loading and inspection of data As shown in Fig. 2 , most workflows begin by loading an experimental dataset of EBSPs. kikuchipy can read files from commercial solutions from Bruker, EDAX, NORDIF, and Oxford Instruments. Before indexing, we recommend getting an overview of the information in the dataset from so-called ‘pre-indexing’ maps. The maps are derived solely from pattern intensities and are independent of any user bias introduced in the indexing setup. They can inform the selection of candidate phases for indexing and help when validating the indexing results. Three such maps are described and used in the first application example. Other maps, such as the ‘quality metrics’ maps suggested by 56 , may also be obtained by applying HyperSpy’s flexible mapping tools. We can inspect patterns by navigating any map with the visualization tools from HyperSpy. Pattern processing Informally, we can interpret an EBSP as the superposition of a diffraction pattern and a

Read the complete text