arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2311.12840cs.CVcs.AIeess.IV
收藏

使用潜在向量表示的晶圆图缺陷模式半监督分类

Wafer Map Defect Patterns Semi-Supervised Classification Using Latent Vector Representation

Shanghai University (SHU)(上海大学)

Qiyu Wei, Wei Zhao, Xiaoyan Zheng, Zeng Zeng

AI总结:

本文提出一种基于预训练 VAE 和 teacher-student network 的晶圆图缺陷模式半监督分类方法,结合故障分布信息指导模型训练,在 WM-811K 数据集上实现了优于现有模型的分类精度与检测性能。

摘要:

随着半导体设计与制造工艺全球化的不断推进,集成电路制造阶段对缺陷检测的需求日益关键,这在提升半导体产品良率方面发挥着重要作用。传统的晶圆图缺陷模式检测方法需要使用电子显微镜收集样本图像进行人工检查,随后由专家评估缺陷。这种方法劳动强度大且效率低下。因此,迫切需要开发一种能够自动检测缺陷的模型以替代人工操作。本文提出一种方法,首先采用预训练的 VAE 模型获取晶圆图的故障分布信息。该信息作为指导,与原始图像集结合用于半监督模型训练。在半监督训练过程中,我们利用 teacher-student network 进行迭代学习。本文提出的模型在基准数据集 WM-811K 晶圆数据集上进行了验证。实验结果表明,与 state-of-the-art 模型相比,该方法具有更优的分类精度和检测性能,满足了工业应用的要求。与原始架构相比,我们取得了显著的性能提升。

英文摘要:

As the globalization of semiconductor design and manufacturing processes continues, the demand for defect detection during integrated circuit fabrication stages is becoming increasingly critical, playing a significant role in enhancing the yield of semiconductor products. Traditional wafer map defect pattern detection methods involve manual inspection using electron microscopes to collect sample images, which are then assessed by experts for defects. This approach is labor-intensive and inefficient. Consequently, there is a pressing need to develop a model capable of automatically detecting defects as an alternative to manual operations. In this paper, we propose a method that initially employs a pre-trained VAE model to obtain the fault distribution information of the wafer map. This information serves as guidance, combined with the original image set for semi-supervised model training. During the semi-supervised training, we utilize a teacher-student network for iterative learning. The model presented in this paper is validated on the benchmark dataset WM-811K wafer dataset. The experimental results demonstrate superior classification accuracy and detection performance compared to state-of-the-art models, fulfilling the requirements for industrial applications. Compared to the original architecture, we have achieved significant performance improvement.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑