Presentation Information
[17p-PA3-1]Estimation of formation states of nano-periodic structures (LIPSS) using scattered-light analysis based on deep learning
〇Shunya Iga1, Yoshio Hayasaki1, Satoshi Hasegawa1 (1.Utsunomiya Univ.)
Keywords:
Laser processing,Nano-periodic structures,Deep learning
LIPSS formed by femtosecond laser irradiation possess useful surface functions, but their formation behavior depends on numerous conditions, making stable formation difficult. This study focuses on the scattered light obtainable during processing and proposes a method using deep learning to predict the presence, periodicity, and orientation of LIPSS from scattered light images. The proposed method was validated by training it to correlate scattered light images with SEM observation results.