基于图像内在正常性的无监督集成电路缺陷分割
Unsupervised Integrated-Circuit Defect Segmentation via Image-Intrinsic Normality
AI总结：
针对IC缺陷分割依赖外部正常集的问题，提出利用图像内在正常性的无监督框架，通过提取正常特征并仅重建正常内容，以残差分割缺陷，在三个工艺阶段数据集上优于现有方法。
摘要：
现代集成电路（IC）制造过程中会产生多种多样、细微的缺陷，这些缺陷降低了良率和可靠性。大多数工业缺陷分割方法将测试图像与外部正常集进行比较，这种策略对于IC图像而言较为脆弱，因为不同产品的布局各异，且精确对齐困难。我们观察到缺陷主要呈局部性，而每幅图像仍包含丰富且可重复的正常模式。因此，我们提出了一种无需外部正常支持的无监督IC缺陷分割框架。一个可学习的正常信息提取器从测试图像中聚合代表性的正常特征，并通过一致性损失强制其与正常区域关联。在这些特征的引导下，解码器仅重建正常内容；重建残差随后用于分割缺陷。伪异常增强进一步稳定了训练过程。在来自三个IC工艺阶段的数据集上的实验表明，该方法相较于现有方法具有一致的改进，并对产品变异性表现出强大的鲁棒性。
英文摘要：
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.
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。
