[2Aa01-14] データ駆動表面科学研究部会「データ駆動アプローチ研究の最近の動向」

[2Aa03] Machine learning applications in multi-dimensional spectromicroscopy

*Naoka Nagamura^1,2,3 (1. National Institute for Materials Science, 2. PRESTO, Japan Science and Technology Agency, 3. Faculty of Advanced Engineering, Tokyo University of Science)

We have investigated local electronic properties of device microstructures during operation by using synchrotron radiation soft X-ray operando scanning photoemission electron microscopy(SR-SPEM).
For accelerating spectral data accumulation and analysis in SPEM measurements, we have developed a high–throughput peak separation method using “Spectrum-adapted ECM algorithm”, and a multi-frame image super-resolution method using sparse modeling.
These machine learning techniques enable us to perform multi-dimensional spectral imaging analysis.