Adaptive partial scanning transmission electron microscopy with reinforcement learning
نویسندگان
چکیده
Abstract Compressed sensing can decrease scanning transmission electron microscopy dose and scan time with minimal information loss. Traditionally, sparse scans used in compressed sample a static set of probing locations. However, dynamic that adapt to specimens are expected be able match or surpass the performance as subset possible scans. Thus, we present prototype for contiguous system piecewise adapts paths they scanned. Sampling directions segments chosen by recurrent neural network (RNN) based on previously observed segments. The RNN is trained reinforcement learning cooperate feedforward convolutional completes This paper presents our policy, experiments, example partial scans, discusses future research directions. Source code, pretrained models, training data openly accessible at https://github.com/Jeffrey-Ede/adaptive-scans .
منابع مشابه
Scanning transmission low-energy electron microscopy
We discuss an extension to the transmission mode of the cathode-lens-equipped scanning electron microscope, enabling operation down to the lowest energies of electrons. Penetration of electrons through free-standing ultrathin films is examined along the full energy scale, and the contribution of the secondary electrons (SEs), released near the bottom surface of the sample, is shown, enhancing t...
متن کاملElemental mapping in scanning transmission electron microscopy
We discuss atomic resolution chemical mapping in scanning transmission electron microscopy (STEM) based on core-loss electron energy loss spectroscopy (EELS) and also on energy dispersive X-ray (EDX) imaging. Chemical mapping using EELS can yield counterintuitive results which, however, can be understood using first principles calculations. Experimental chemical maps based on EDX bear out the t...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Machine learning: science and technology
سال: 2021
ISSN: ['2632-2153']
DOI: https://doi.org/10.1088/2632-2153/abf5b6