Defect detection in photolithographic patterns using deep learning models trained on synthetic data

In the photolithographic process vital to semiconductor manufacturing, various types of defects appear during EUV pattering. Due to ever-shrinking pattern size, these defects are extremely small and cause false or missed detection during inspection. Specifically, the lack of defect-annotated quality...

Full description

Saved in:
Bibliographic Details
Main Authors: Prashant P. Shinde, Priyadarshini P. Pai, Shashishekar P. Adiga, K. Subramanya Mayya, Yongbeom Seo, Myungsoo Hwang, Heeyoung Go, Changmin Park
Format: Article
Language:English
Published: Elsevier 2025-05-01
Series:Heliyon
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S240584402501761X
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850123518190026752
author Prashant P. Shinde
Priyadarshini P. Pai
Shashishekar P. Adiga
K. Subramanya Mayya
Yongbeom Seo
Myungsoo Hwang
Heeyoung Go
Changmin Park
author_facet Prashant P. Shinde
Priyadarshini P. Pai
Shashishekar P. Adiga
K. Subramanya Mayya
Yongbeom Seo
Myungsoo Hwang
Heeyoung Go
Changmin Park
author_sort Prashant P. Shinde
collection DOAJ
description In the photolithographic process vital to semiconductor manufacturing, various types of defects appear during EUV pattering. Due to ever-shrinking pattern size, these defects are extremely small and cause false or missed detection during inspection. Specifically, the lack of defect-annotated quality data with good representation of smaller defects has prohibited deployment of deep learning based defect detection models in fabrication lines. To resolve the problem of data unavailability, we artificially generate scanning electron microscopy (SEM) images of line patterns with known distribution of defects and autonomously annotate them. We then employ state-of-the-art object detection models to investigate defect detection performance as a function of defect size, much smaller than the pitch width. We find that the real-time object detector YOLOv8 has the best mean average precision of 96% as compared to EfficientNet, 83%, and SSD, 77%, with the ability to detect smaller defects. We report the smallest defect size that can be detected reliably. When tested on real SEM data, the YOLOv8 model correctly detected 84.6% of Bridge defects and 78.3% of Break defects across all relevant instances. These promising results suggest that synthetic data can be used as an alternative to real-world data in order to develop robust machine-learning models.
format Article
id doaj-art-278f913628e844a88f3f436b3ba83ee9
institution OA Journals
issn 2405-8440
language English
publishDate 2025-05-01
publisher Elsevier
record_format Article
series Heliyon
spelling doaj-art-278f913628e844a88f3f436b3ba83ee92025-08-20T02:34:35ZengElsevierHeliyon2405-84402025-05-011110e4337710.1016/j.heliyon.2025.e43377Defect detection in photolithographic patterns using deep learning models trained on synthetic dataPrashant P. Shinde0Priyadarshini P. Pai1Shashishekar P. Adiga2K. Subramanya Mayya3Yongbeom Seo4Myungsoo Hwang5Heeyoung Go6Changmin Park7NextGen Projects (SAIT-India), Samsung Semiconductor India Research, Bangalore, 560048, India; Corresponding authors.NextGen Projects (SAIT-India), Samsung Semiconductor India Research, Bangalore, 560048, IndiaNextGen Projects (SAIT-India), Samsung Semiconductor India Research, Bangalore, 560048, India; Corresponding authors.NextGen Projects (SAIT-India), Samsung Semiconductor India Research, Bangalore, 560048, IndiaFoundry Process Development Team, Semiconductor R&D Center, Samsung Electronics, Seoul, KoreaFoundry Process Development Team, Semiconductor R&D Center, Samsung Electronics, Seoul, KoreaFoundry Process Development Team, Semiconductor R&D Center, Samsung Electronics, Seoul, KoreaFoundry Process Development Team, Semiconductor R&D Center, Samsung Electronics, Seoul, KoreaIn the photolithographic process vital to semiconductor manufacturing, various types of defects appear during EUV pattering. Due to ever-shrinking pattern size, these defects are extremely small and cause false or missed detection during inspection. Specifically, the lack of defect-annotated quality data with good representation of smaller defects has prohibited deployment of deep learning based defect detection models in fabrication lines. To resolve the problem of data unavailability, we artificially generate scanning electron microscopy (SEM) images of line patterns with known distribution of defects and autonomously annotate them. We then employ state-of-the-art object detection models to investigate defect detection performance as a function of defect size, much smaller than the pitch width. We find that the real-time object detector YOLOv8 has the best mean average precision of 96% as compared to EfficientNet, 83%, and SSD, 77%, with the ability to detect smaller defects. We report the smallest defect size that can be detected reliably. When tested on real SEM data, the YOLOv8 model correctly detected 84.6% of Bridge defects and 78.3% of Break defects across all relevant instances. These promising results suggest that synthetic data can be used as an alternative to real-world data in order to develop robust machine-learning models.http://www.sciencedirect.com/science/article/pii/S240584402501761XScanning electron microscopyData generationMachine learningDefect detection
spellingShingle Prashant P. Shinde
Priyadarshini P. Pai
Shashishekar P. Adiga
K. Subramanya Mayya
Yongbeom Seo
Myungsoo Hwang
Heeyoung Go
Changmin Park
Defect detection in photolithographic patterns using deep learning models trained on synthetic data
Heliyon
Scanning electron microscopy
Data generation
Machine learning
Defect detection
title Defect detection in photolithographic patterns using deep learning models trained on synthetic data
title_full Defect detection in photolithographic patterns using deep learning models trained on synthetic data
title_fullStr Defect detection in photolithographic patterns using deep learning models trained on synthetic data
title_full_unstemmed Defect detection in photolithographic patterns using deep learning models trained on synthetic data
title_short Defect detection in photolithographic patterns using deep learning models trained on synthetic data
title_sort defect detection in photolithographic patterns using deep learning models trained on synthetic data
topic Scanning electron microscopy
Data generation
Machine learning
Defect detection
url http://www.sciencedirect.com/science/article/pii/S240584402501761X
work_keys_str_mv AT prashantpshinde defectdetectioninphotolithographicpatternsusingdeeplearningmodelstrainedonsyntheticdata
AT priyadarshinippai defectdetectioninphotolithographicpatternsusingdeeplearningmodelstrainedonsyntheticdata
AT shashishekarpadiga defectdetectioninphotolithographicpatternsusingdeeplearningmodelstrainedonsyntheticdata
AT ksubramanyamayya defectdetectioninphotolithographicpatternsusingdeeplearningmodelstrainedonsyntheticdata
AT yongbeomseo defectdetectioninphotolithographicpatternsusingdeeplearningmodelstrainedonsyntheticdata
AT myungsoohwang defectdetectioninphotolithographicpatternsusingdeeplearningmodelstrainedonsyntheticdata
AT heeyounggo defectdetectioninphotolithographicpatternsusingdeeplearningmodelstrainedonsyntheticdata
AT changminpark defectdetectioninphotolithographicpatternsusingdeeplearningmodelstrainedonsyntheticdata