[22p-M206-1~16] 23.1 Joint Session N "Informatics"

[22p-M206-12] Improvement of Electron Holography Precision Using Direct Detection Camera and Sparse Coding

〇Satoshi Anada¹, Yuki Nomura¹, Tsukasa Hirayama¹, Kazuo Yamamoto¹ (1.JFCC)

Keywords:machine learning, sparse coding, electron holography

Precise assessment of nanometer-scale electromagnetic distribution is frequently required to study and develop materials and devices. In our study, we have concluded that an electron direct detector (high-sensitivity CMOS camera) and a machine learning method called sparse coding can improve the precition of electromagnetic measurement using electron holography to achieve 20 times of the conventional method using a CCD camera without sparse coding.