[21p-B205-1~8] 3.6 Laser processing (formerly 3.7)

1:30 PM - 1:45 PM JST(4:30 AM - 4:45 AM UTC)

[21p-B205-1] Detection of crack formation from the optical transmission image in laser grooving of glasses using deep neural network

〇Dai Yoshitomi¹, Hiroshi Ogawa¹, Daisuke Sato¹, Hideyuki Takada¹, Masahito Tanaka¹, Ryunosuke Kuroda¹, Aiko Narazaki¹ (1.AIST)

Keywords:

laser machining,deep neural network,crack

In the laser machining of glasses, it is crucial to suppress the crack formation, because it will reduce the machining accuracy and/or cause the mechanical damage. In this work, we demonstrate the discrimination of crack formation in ultrafast laser grooving of silica glasses from the optical transmission image by utilizing the deep neural network.