YOLOv8 for Defect Inspection of Hexagonal Directed Self-Assembly Patterns: A Data-Centric Approach

TL;DR

This paper introduces a data-centric approach using YOLOv8 for defect inspection in hexagonal DSA patterns, emphasizing high-quality labeling with minimal expert effort, achieving over 0.9 mAP in defect detection.

Contribution

It presents a novel labeling method for DSA pattern datasets and demonstrates YOLOv8's effectiveness in defect detection with high precision.

Findings

YOLOv8 achieves >0.9 mAP on DSA defect dataset.

A minimal-effort labeling approach produces high-quality annotations.

The dataset reflects expert defect labeling expectations.

Abstract

Shrinking pattern dimensions leads to an increased variety of defect types in semiconductor devices. This has spurred innovation in patterning approaches such as Directed self-assembly (DSA) for which no traditional, automatic defect inspection software exists. Machine Learning-based SEM image analysis has become an increasingly popular research topic for defect inspection with supervised ML models often showing the best performance. However, little research has been done on obtaining a dataset with high-quality labels for these supervised models. In this work, we propose a method for obtaining coherent and complete labels for a dataset of hexagonal contact hole DSA patterns while requiring minimal quality control effort from a DSA expert. We show that YOLOv8, a state-of-the-art neural network, achieves defect detection precisions of more than 0.9 mAP on our final dataset which best reflects DSA expert defect labeling expectations. We discuss the strengths and limitations of our proposed labeling approach and suggest directions for future work in data-centric ML-based defect inspection.