[7p-S202-1~12] 23.1 Joint Session N "Informatics"

2:45 PM - 3:00 PM JST(5:45 AM - 6:00 AM UTC)

[7p-S202-7] Deep Learning-Based Automated Evaluation Pipeline for Semiconductor Etched Geometry

〇Seiyo Nojima¹, Kazunori Iwamitsu¹, Zentaro Akase¹, Akihiko Kikuchi², Shigetaka Tomiya¹ (1.NAIST, 2.Sophia Univ.)

Keywords:

deep learning,semiconductor,etching

In semiconductor devices, the dimensions and shapes of microstructures significantly affect device performance, making accurate measurement using electron microscopy essential. However, manual evaluation of electron microscope images is time-consuming and susceptible to operator bias. Therefore, automated evaluation using deep learning is highly anticipated. In this study, we aimed to develop an automated evaluation pipeline capable of handling diverse images, and implemented multiple configurations to compare the robustness of image analysis.