Using cross-correlation for automated stitching of two-dimensional multi-tile electron backscatter diffraction data.

Area of Science:

• Materials Science
• Crystallography
• Electron Microscopy

Background:

• Accurate alignment of large, multi-tile datasets is crucial for detailed microstructural analysis.
• Existing methods for stitching electron backscatter diffraction (EBSD) data can be time-consuming and lack high accuracy.

Purpose of the Study:

• To develop an automated, high-accuracy method for aligning consecutive two-dimensional multi-tile EBSD scans.
• To improve the stitching process for large-scale EBSD datasets.

Main Methods:

• Locating grain and phase boundaries in overlapping regions of adjacent scan tiles.
• Employing cross-correlation algorithms to determine optimal relative positions of scan tiles.
• Utilizing 2D Savitzky-Golay filtering for background estimation and subtraction to enhance signal in high-interface-density samples.

Main Results:

• Demonstrated the technique's effectiveness on datasets with varying interface densities.
• Achieved high accuracy in automatically aligning multi-tile EBSD scans.
• Successfully implemented the method as enhancements to open-source EBSD stitching software.

Conclusions:

• The developed method provides an accurate and automated solution for stitching large EBSD datasets.
• This advancement facilitates more detailed and reliable microstructural characterization.
• The integration into open-source code promotes wider accessibility and application in materials research.