An Efficient Transformer-Based Approach for DUV Lithography SEM Image Denoising

  • Jiwei Shen
  • , Botong Zhao
  • , Hu Lu
  • , Pengjie Lou
  • , Wenzhan Zhou
  • , Kan Zhou
  • , Xintong Zhao
  • , Shujing Lyu
  • , Yue Lu*
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Scopus citations

Abstract

Traditional Scanning Electron Microscopy (SEM) image noise reduction techniques, such as frame averaging or utilizing higher resolution SEM images, may result in potential electron beam damage and could also limit the speed of screening. In this paper, we propose a deep-learning-based denoising method using a Transformer-based architecture that addresses these challenges. This method effectively reduces noise while preserving image details, providing comparable measurements such as line width roughness that are only attainable with higher signal-to-noise ratio SEM images.

Original languageEnglish
Title of host publicationIWAPS 2023 - 2023 7th International Workshop on Advanced Patterning Solutions
EditorsYayi Wei, Tianchun Ye
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798350344547
DOIs
StatePublished - 2023
Event7th International Workshop on Advanced Patterning Solutions, IWAPS 2023 - Lishui, Zhejiang Province, China
Duration: 26 Oct 202327 Oct 2023

Publication series

NameIWAPS 2023 - 2023 7th International Workshop on Advanced Patterning Solutions

Conference

Conference7th International Workshop on Advanced Patterning Solutions, IWAPS 2023
Country/TerritoryChina
CityLishui, Zhejiang Province
Period26/10/2327/10/23

Keywords

  • Lithography
  • SEM Image Denoising
  • Transformer
  • deep learning

Fingerprint

Dive into the research topics of 'An Efficient Transformer-Based Approach for DUV Lithography SEM Image Denoising'. Together they form a unique fingerprint.

Cite this