Unsupervised Multi-Clustering and Decision-Making Strategies for 4D-STEM Orientation Mapping

Abstract

This study presents a novel integration of unsupervised learning and decision-making strategies for the advanced analysis of 4D-STEM datasets, with a focus on non-negative matrix factorization (NMF) as the primary clustering method. Our approach introduces a systematic framework to determine the optimal number of components (k) required for robust and interpretable orientation mapping. By leveraging the K-Component Loss method and Image Quality Assessment (IQA) metrics, we effectively balance reconstruction fidelity and model complexity. Additionally, we highlight the critical role of dataset preprocessing in improving clustering stability and accuracy. Furthermore, our spatial weight matrix analysis provides insights into overlapping regions within the dataset by employing threshold-based visualization, facilitating a detailed understanding of cluster interactions. The results demonstrate the potential of combining NMF with advanced IQA metrics and preprocessing techniques for reliable orientation mapping and structural analysis in 4D-STEM datasets, paving the way for future applications in multi-dimensional material characterization.

32 pages, 5 figures, 5 figures in SI

References in corpus (6)

• Gradient Magnitude Similarity Deviation: A Highly Efficient Perceptual Image Quality Index
• py4DSTEM: a software package for multimodal analysis of four-dimensional scanning transmission electron microscopy datasets
• Mean Deviation Similarity Index: Efficient and Reliable Full-Reference Image Quality Evaluator
• Fast Grain Mapping with Sub-Nanometer Resolution Using 4D-STEM with Grain Classification by Principal Component Analysis and Non-Negative Matrix Factorization
• Phase Object Reconstruction for 4D-STEM using Deep Learning
• Improved ACOM pattern matching in 4D STEM through adaptive sub pixel peak detection and image reconstruction