Convolutional Neural Network for Wafer Surface Defect Classification and the Detection of Unknown Defect Class

Sejune Cheon, Hankang Lee, Chang Ouk Kim, Seok Hyung Lee

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

An automatic defect classification (ADC) system identifies and classifies wafer surface defects using scanning electron microscope images. By classifying defects, manufacturers can determine whether the wafer can be repaired and proceed to the next fabrication step. Current ADC systems have high defect detection performance. However, the classification power is poor. In most work sites, defect classification is performed manually using the naked eye, which is unreliable. This paper proposes an ADC method based on deep learning that automatically classifies various types of wafer surface damage. In contrast to conventional ADC methods, which apply a series of image recognition and machine learning techniques to find features for defect classification, the proposed model adopts a single convolutional neural network (CNN) model that can extract effective features for defect classification without using additional feature extraction algorithms. Moreover, the proposed method can identify defect classes not seen during training by comparing the CNN features of the unseen classes with those of the trained classes. Experiments with real datasets verified that the proposed ADC method achieves high defect classification performance.

Original languageEnglish
Article number8657760
Pages (from-to)163-170
Number of pages8
JournalIEEE Transactions on Semiconductor Manufacturing
Volume32
Issue number2
DOIs
Publication statusPublished - 2019 May

Fingerprint

Surface defects
surface defects
wafers
Neural networks
Defects
defects
learning
Image recognition
machine learning
classifying
pattern recognition
Learning systems
Feature extraction
Electron microscopes
education
electron microscopes
Scanning
Fabrication
damage

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Industrial and Manufacturing Engineering
  • Electrical and Electronic Engineering

Cite this

@article{aa941d12028646c7a213bf3e6f84b3ec,
title = "Convolutional Neural Network for Wafer Surface Defect Classification and the Detection of Unknown Defect Class",
abstract = "An automatic defect classification (ADC) system identifies and classifies wafer surface defects using scanning electron microscope images. By classifying defects, manufacturers can determine whether the wafer can be repaired and proceed to the next fabrication step. Current ADC systems have high defect detection performance. However, the classification power is poor. In most work sites, defect classification is performed manually using the naked eye, which is unreliable. This paper proposes an ADC method based on deep learning that automatically classifies various types of wafer surface damage. In contrast to conventional ADC methods, which apply a series of image recognition and machine learning techniques to find features for defect classification, the proposed model adopts a single convolutional neural network (CNN) model that can extract effective features for defect classification without using additional feature extraction algorithms. Moreover, the proposed method can identify defect classes not seen during training by comparing the CNN features of the unseen classes with those of the trained classes. Experiments with real datasets verified that the proposed ADC method achieves high defect classification performance.",
author = "Sejune Cheon and Hankang Lee and Kim, {Chang Ouk} and Lee, {Seok Hyung}",
year = "2019",
month = "5",
doi = "10.1109/TSM.2019.2902657",
language = "English",
volume = "32",
pages = "163--170",
journal = "IEEE Transactions on Semiconductor Manufacturing",
issn = "0894-6507",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "2",

}

Convolutional Neural Network for Wafer Surface Defect Classification and the Detection of Unknown Defect Class. / Cheon, Sejune; Lee, Hankang; Kim, Chang Ouk; Lee, Seok Hyung.

In: IEEE Transactions on Semiconductor Manufacturing, Vol. 32, No. 2, 8657760, 05.2019, p. 163-170.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Convolutional Neural Network for Wafer Surface Defect Classification and the Detection of Unknown Defect Class

AU - Cheon, Sejune

AU - Lee, Hankang

AU - Kim, Chang Ouk

AU - Lee, Seok Hyung

PY - 2019/5

Y1 - 2019/5

N2 - An automatic defect classification (ADC) system identifies and classifies wafer surface defects using scanning electron microscope images. By classifying defects, manufacturers can determine whether the wafer can be repaired and proceed to the next fabrication step. Current ADC systems have high defect detection performance. However, the classification power is poor. In most work sites, defect classification is performed manually using the naked eye, which is unreliable. This paper proposes an ADC method based on deep learning that automatically classifies various types of wafer surface damage. In contrast to conventional ADC methods, which apply a series of image recognition and machine learning techniques to find features for defect classification, the proposed model adopts a single convolutional neural network (CNN) model that can extract effective features for defect classification without using additional feature extraction algorithms. Moreover, the proposed method can identify defect classes not seen during training by comparing the CNN features of the unseen classes with those of the trained classes. Experiments with real datasets verified that the proposed ADC method achieves high defect classification performance.

AB - An automatic defect classification (ADC) system identifies and classifies wafer surface defects using scanning electron microscope images. By classifying defects, manufacturers can determine whether the wafer can be repaired and proceed to the next fabrication step. Current ADC systems have high defect detection performance. However, the classification power is poor. In most work sites, defect classification is performed manually using the naked eye, which is unreliable. This paper proposes an ADC method based on deep learning that automatically classifies various types of wafer surface damage. In contrast to conventional ADC methods, which apply a series of image recognition and machine learning techniques to find features for defect classification, the proposed model adopts a single convolutional neural network (CNN) model that can extract effective features for defect classification without using additional feature extraction algorithms. Moreover, the proposed method can identify defect classes not seen during training by comparing the CNN features of the unseen classes with those of the trained classes. Experiments with real datasets verified that the proposed ADC method achieves high defect classification performance.

UR - http://www.scopus.com/inward/record.url?scp=85065425987&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85065425987&partnerID=8YFLogxK

U2 - 10.1109/TSM.2019.2902657

DO - 10.1109/TSM.2019.2902657

M3 - Article

AN - SCOPUS:85065425987

VL - 32

SP - 163

EP - 170

JO - IEEE Transactions on Semiconductor Manufacturing

JF - IEEE Transactions on Semiconductor Manufacturing

SN - 0894-6507

IS - 2

M1 - 8657760

ER -