1. Identification of crystal symmetry from noisy diffraction patterns by a shape analysis and deep learning
- Author
-
Donghun Kim, Jeongrae Kim, Sang Soo Han, and Leslie Ching Ow Tiong
- Subjects
FOS: Computer and information sciences ,Diffraction ,Computer Science - Machine Learning ,Computer science ,Crystal system ,FOS: Physical sciences ,02 engineering and technology ,Crystal structure ,010402 general chemistry ,01 natural sciences ,Convolutional neural network ,Machine Learning (cs.LG) ,QA76.75-76.765 ,General Materials Science ,Computer software ,Materials of engineering and construction. Mechanics of materials ,business.industry ,Deep learning ,Space group ,Pattern recognition ,Computational Physics (physics.comp-ph) ,021001 nanoscience & nanotechnology ,0104 chemical sciences ,Computer Science Applications ,Mechanics of Materials ,Modeling and Simulation ,TA401-492 ,Artificial intelligence ,0210 nano-technology ,business ,Physics - Computational Physics ,Shape analysis (digital geometry) ,Monoclinic crystal system - Abstract
The robust and automated determination of crystal symmetry is of utmost importance in material characterization and analysis. Recent studies have shown that deep learning (DL) methods can effectively reveal the correlations between X-ray or electron-beam diffraction patterns and crystal symmetry. Despite their promise, most of these studies have been limited to identifying relatively few classes into which a target material may be grouped. On the other hand, the DL-based identification of crystal symmetry suffers from a drastic drop in accuracy for problems involving classification into tens or hundreds of symmetry classes (e.g., up to 230 space groups), severely limiting its practical usage. Here, we demonstrate that a combined approach of shaping diffraction patterns and implementing them in a multistream DenseNet (MSDN) substantially improves the accuracy of classification. Even with an imbalanced dataset of 108,658 individual crystals sampled from 72 space groups, our model achieves 80.12 ± 0.09% space group classification accuracy, outperforming conventional benchmark models by 17–27 percentage points (%p). The enhancement can be largely attributed to the pattern shaping strategy, through which the subtle changes in patterns between symmetrically close crystal systems (e.g., monoclinic vs. orthorhombic or trigonal vs. hexagonal) are well differentiated. We additionally find that the MSDN architecture is advantageous for capturing patterns in a richer but less redundant manner relative to conventional convolutional neural networks. The proposed protocols in regard to both input descriptor processing and DL architecture enable accurate space group classification and thus improve the practical usage of the DL approach in crystal symmetry identification.
- Published
- 2020