[论文解读] SAMSEM -- A Generic and Scalable Approach for IC Metal Line Segmentation

Christian Gehrmann, Jonas Ricker+6 more|arXiv (Cornell University)|Mar 17, 2026

Physical Unclonable Functions (PUFs) and Hardware Security被引用 0

一句话总结

SAMSEM 将 SAM2 应用于 IC 金属层 SEM 图像，采用多尺度分割和基于拓扑的损失以在未见 IC 上实现更低错误率的泛化。其在分布内错误率为 0.72%，分布外错误率为 5.53%，在使用全部数据训练时降至 0.62%。

ABSTRACT

In light of globalized hardware supply chains, the assurance of hardware components has gained significant interest, particularly in cryptographic applications and high-stakes scenarios. Identifying metal lines on scanning electron microscope (SEM) images of integrated circuits (ICs) is one essential step in verifying the absence of malicious circuitry in chips manufactured in untrusted environments. Due to varying manufacturing processes and technologies, such verification usually requires tuning parameters and algorithms for each target IC. Often, a machine learning model trained on images of one IC fails to accurately detect metal lines on other ICs. To address this challenge, we create SAMSEM by adapting Meta's Segment Anything Model 2 (SAM2) to the domain of IC metal line segmentation. Specifically, we develop a multi-scale segmentation approach that can handle SEM images of varying sizes, resolutions, and magnifications. Furthermore, we deploy a topology-based loss alongside pixel-based losses to focus our segmentation on electrical connectivity rather than pixel-level accuracy. Based on a hyperparameter optimization, we then fine-tune the SAM2 model to obtain a model that generalizes across different technology nodes, manufacturing materials, sample preparation methods, and SEM imaging technologies. To this end, we leverage an unprecedented dataset of SEM images obtained from 48 metal layers across 14 different ICs. When fine-tuned on seven ICs, SAMSEM achieves an error rate as low as 0.72% when evaluated on other images from the same ICs. For the remaining seven unseen ICs, it still achieves error rates as low as 5.53%. Finally, when fine-tuned on all 14 ICs, we observe an error rate of 0.62%. Hence, SAMSEM proves to be a reliable tool that significantly advances the frontier in metal line segmentation, a key challenge in post-manufacturing IC verification.

研究动机与目标

• 解决在不同 IC 和 SEM 条件下对金属线分割的泛化挑战，而无需手动重新调参。
• 开发一个鲁棒、可扩展的流水线，处理不同的图像尺寸、放大倍数和层结构。
• 优先考虑电气连通性而非像素级精确性，以改善下游网表提取。

提出的方法

• 在大型 SEM 金属层数据集上对 Meta 的 Segment Anything Model 2 (SAM2) 进行微调。
• 实现一个多尺度分割流水线，结合全图和 512×512 小块，以处理不同结构大小。
• 在像素级损失之外引入基于拓扑的损失，强调正确的电气连通性。
• 使用超参数优化和数据增强以提高在不同工艺节点和 SEM 设置下的泛化能力。
• 提供一个决策机制，在patch级最终分割时在模型输出之间进行选择。

实验结果

研究问题

• RQ1 一个经过微调的基于 SAM2 的模型能在未见 IC 和不同金属层的 SEM 图像中实现多大程度的泛化？
• RQ2 多尺度分割方法是否能在不同分辨率和放大倍数下保持高分割质量？
• RQ3 在 IC 金属线分割中，基于拓扑的损失是否比仅像素级损失更有效地降低连通性错误（短路/开路）？
• RQ4 对不同 SAM2 组件（图像编码器、掩码解码器、提示编码器）的微调对分割性能有何影响？
• RQ5 哪些数据集和训练策略能实现跨 IC 的金属线分割的鲁棒泛化而无需手动重新调参？

主要发现

• 跨 IC 泛化强：在对七个未见 IC 进行微调时，分布外错误率降至 5.53%；在用全部 14 个 IC 训练时，分布内错误率降至 0.62%。
• 所提的多尺度分割（全尺寸 + 512×512 小块）在处理大尺度和细粒度金属结构上均有改进。
• 聚焦于电气连通性的拓扑损失减少短路/开路错误，并提升网表可靠性，优于仅像素级损失。
• 将图像编码器、掩码解码器和提示编码器共同微调在 Ablations 中表现最好，分布内 ESD 误差为 0.7%。
• 该方法在覆盖从 200 nm 到 20 nm 的工艺节点、14 个 IC 和 48 条金属层的多样数据集上得到验证。