Overcoming Data Scarcity and Confidentiality in Hardware Assurance via Synthetic Generation

论文概要

研究领域: CV 作者: Gijung Lee, Ronald Wilson, Damon L. Woodard 发布时间: 2026-08-11 arXiv: 2508.03801

中文摘要

硬件保证依赖扫描电子显微镜（SEM）来验证纳米级结构，但组装自动化分析所需的大规模、高质量数据集受到耗时采集和专有设计严格知识产权（IP）限制的双重阻碍。我们提出了一种隐私保护流程，通过对功能设计进行重度扭曲来保护IP，同时从少量初始示例生成视觉上逼真的合成数据集。StyleGAN首先学习硬件布局掩模的分布，以生成新颖的、宏观上多样化的结构。随后，条件GAN（Pix2PixHD）将这些掩模转换为保留真实纹理和噪声的逼真SEM图像。这项工作的主要发现是，仅在这种合成数据上训练的分割模型不仅展示了成功的「仿真到真实」迁移到真实图像，而且优于在有限真实数据集上训练的基线模型。因为底层合成布局明显是新颖的，并且不复制原始设计的任何特定专有布线，部署最终的分割模型减轻了将敏感IP暴露于梯度反转和成员推断等攻击的风险，为硬件保证提供了一个高度安全、高性能的解决方案。

原文摘要

Hardware assurance relies on scanning electron microscopy (SEM) to verify nanoscale structures, but assembling the large, high-quality datasets required for automated analysis is impeded by time-intensive acquisition and strict intellectual property (IP) constraints on proprietary designs. We propose a privacy-preserving pipeline that secures IP by heavily distorting the functional design while generating a visually realistic synthetic dataset from a small set of initial examples. A StyleGAN first learns the distribution of hardware layout masks to generate novel, macroscopically varied structures. Subsequently, a conditional GAN (Pix2PixHD) translates these masks into realistic SEM images that preserve authentic textures and noise. The primary finding of this work is that a segmentation m...

--- *自动采集于 2026-08-12*

#论文 #arXiv #CV #小凯