Skip to content

mizign/ML-MoS2-defect-detect

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 

Repository files navigation

ML-MoS2-defect-detect

MoS2 (이황화 몰리브데늄)을 촬영한 STEM 이미지상에 나타난 Sulfur defects를 감지하는 머신러닝 모델을 제작했습니다.

🔗 레퍼런스: STEM Image Analysis Based on Deep Learning: Identification of Vacancy Defects and Polymorphs of MoS2

🔗 레퍼런스: Generating STEM Simulated Image


Introduction

image image


Architecture

image image


Generating Training Set

image


Training

image


Crop Augmentation의 효과: Augmentation I vs Augmentation II

  • Augmentation I: w/ Crop Augmentation
  • Augmentation II w/o Crop Augmentation image

Case1: Predict Region: 5x5nm

image\


Case2: Predict Region: 2.5x2.5nm

crop augmentation 과정을 거치지 않은 Training Set으로 훈련시킨 모델의 경우 2.5x2.5nm 영역에 대한 예측성능이 상대적으로 낮다는 것을 확인할 수 있었습니다.
image

About

MoS2을 촬영한 STEM 이미지상에 나타난 Sulfur defects를 감지하는 머신러닝 모델의 제작

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors