无需外部正常样本，通过图像内在规律实现芯片缺陷无监督分割
Unsupervised Integrated-Circuit Defect Segmentation via Image-Intrinsic Normality
- 从测试图自身提取正常特征，构建自洽的正常模式
- 重建残差定位缺陷，三阶段数据集上性能超越现有方法
- 适合跨产品型号的芯片制造缺陷检测场景
现代集成电路（IC）制造中存在多种细粒度缺陷，影响良率与可靠性。现有工业缺陷分割通常将待测图像与外部正常样本对比，但在芯片图像中因版图差异大、对齐困难，该方法脆弱。我们观察到缺陷多为局部异常，而每幅图像仍包含丰富且可重复的正常纹理模式。因此提出一种无需外部正常样本支持的无监督芯片缺陷分割框架：可学习的正常信息提取器从测试图像中聚合代表性正常特征，相干性损失强制其与正常区域关联；解码器仅重构正常内容，重建残差即用于缺陷分割。伪异常增强进一步稳定训练。在三个工艺阶段的数据集上实验表明，该方法性能持续优于现有方法，且对产品变异性具有强鲁棒性。
原文摘要 · Abstract (English)
Modern Integrated-Circuit(IC) manufacturing introduces diverse, fine-grained defects that depress yield and reliability. Most industrial defect segmentation compares a test image against an external normal set, a strategy that is brittle for IC imagery where layouts vary across products and accurate alignment is difficult. We observe that defects are predominantly local, while each image still contains rich, repeatable normal patterns. We therefore propose an unsupervised IC defect segmentation framework that requires no external normal support. A learnable normal-information extractor aggregates representative normal features from the test image, and a coherence loss enforces their association with normal regions. Guided by these features, a decoder reconstructs only normal content; the reconstruction residual then segments defects. Pseudo-anomaly augmentation further stabilizes training. Experiments on datasets from three IC process stages show consistent improvements over existing approaches and strong robustness to product variability.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。