A Lightweight Transformer for Faster and Robust EBSD Data Collection
This work addresses data collection challenges in materials science, offering a robust and faster solution for 3D EBSD, though it is incremental as it builds on existing transformer architectures.
The paper tackles the problem of fluctuating data quality and slow collection in 3D EBSD microscopy by introducing a two-step method using a lightweight transformer and projection algorithm to recover missing slices, achieving superior recovery accuracy on real data compared to existing methods.
Three dimensional electron back-scattered diffraction (EBSD) microscopy is a critical tool in many applications in materials science, yet its data quality can fluctuate greatly during the arduous collection process, particularly via serial-sectioning. Fortunately, 3D EBSD data is inherently sequential, opening up the opportunity to use transformers, state-of-the-art deep learning architectures that have made breakthroughs in a plethora of domains, for data processing and recovery. To be more robust to errors and accelerate this 3D EBSD data collection, we introduce a two step method that recovers missing slices in an 3D EBSD volume, using an efficient transformer model and a projection algorithm to process the transformer's outputs. Overcoming the computational and practical hurdles of deep learning with scarce high dimensional data, we train this model using only synthetic 3D EBSD data with self-supervision and obtain superior recovery accuracy on real 3D EBSD data, compared to existing methods.