[19p-B2-1~14] 3.6 Laser processing

1:15 PM - 1:30 PM JST(4:15 AM - 4:30 AM UTC)

[19p-B2-2] Structure estimation based on scattered light image of nano-periodic structure
using deep learning

〇(M2)Ryota Masuda¹, Yoshio Hayasaki¹, Satoshi Hasegawa¹ (1.Center for Optical Research and Education, Utsunomiya Univ.)

Keywords:

nano-periodic structure,deep learning

For stable formation of nanostructures formed by femtosecond laser pulse irradiation, it is necessary to observe the nanostructures in-situ during laser processing and adjust the conditions. However, since optical methods cannot observe the nanostructures, SEM is used, but in-situ observation is difficult because a vacuum environment is required. In this study, we verified that SEM images of nano-periodic structures formed on silicon substrates can be recovered from 2D Fourier transform images using a deep learning model for in-situ measurement.