用于工业异常检测的组件感知无监督逻辑异常生成
Component-aware Unsupervised Logical Anomaly Generation for Industrial Anomaly Detection
AI总结：
本文提出组件感知无监督框架 ComGEN，通过多组件解耦与注意力引导残差编辑生成逻辑异常，在 MVTecLOCO 上达到 91.2% AUROC，并提升真实工业场景检测性能。
摘要：
异常检测在工业制造中对于确保产品质量和提升自动化流程效率至关重要。异常样本的稀缺限制了传统检测方法，因此异常生成对于扩充数据库至关重要。然而，近期的生成模型往往产生不切实际的异常，从而增加误报，或者需要真实世界的异常样本来进行训练。在本工作中，我们将异常生成视为一个组合问题，并提出 ComGEN，这是一个组件感知且无监督的框架，用于弥补逻辑异常生成方面的空白。我们的方法包括用于解耦视觉组件的多组件学习策略，随后是生成编辑流程。解耦的文本到组件对揭示内在逻辑约束，执行注意力引导的残差映射，并在多个尺度上使用迭代匹配的参考进行模型训练。在 MVTecLOCO 数据集上的实验证实了 ComGEN 的有效性，其取得了 91.2% 的最佳 AUROC 分数。在真实世界柴油发动机场景和广泛使用的 MVTecAD 数据集上的额外实验表明，将 ComGEN 生成的模拟异常集成到自动化生产工作流中时，性能获得了显著提升。
英文摘要：
Anomaly detection is critical in industrial manufacturing for ensuring product quality and improving efficiency in automated processes. The scarcity of anomalous samples limits traditional detection methods, making anomaly generation essential for expanding the data repository. However, recent generative models often produce unrealistic anomalies increasing false positives, or require real-world anomaly samples for training. In this work, we treat anomaly generation as a compositional problem and propose ComGEN, a component-aware and unsupervised framework that addresses the gap in logical anomaly generation. Our method comprises a multi-component learning strategy to disentangle visual components, followed by subsequent generation editing procedures. Disentangled text-to-component pairs, revealing intrinsic logical constraints, conduct attention-guided residual mapping and model training with iteratively matched references across multiple scales. Experiments on the MVTecLOCO dataset confirm the efficacy of ComGEN, achieving the best AUROC score of 91.2%. Additional experiments on the real-world scenario of Diesel Engine and widely-used MVTecAD dataset demonstrate significant performance improvements when integrating simulated anomalies generated by ComGEN into automated production workflows.
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。
