To improve logical anomaly detection, some previous works have integrated segmentation techniques with conventional anomaly detection methods. Although these methods are effective, they frequently lead to unsatisfactory segmentation results and require manual annotations. To address these drawbacks, we develop an unsupervised component segmentation technique that leverages foundation models to autonomously generate training labels for a lightweight segmentation network without human labeling. Integrating this new segmentation technique with our proposed Patch Histogram module and the Local-Global Student-Teacher (LGST) module, we achieve a detection AUROC of 95.3% in the MVTec LOCO AD dataset, which surpasses previous SOTA methods. Furthermore, our proposed method provides lower latency and higher throughput than most existing approaches.

翻译：为提升逻辑异常检测性能，部分先前研究将分割技术与传统异常检测方法相结合。尽管这些方法有效，但常导致分割结果不尽如人意且需人工标注。针对这些缺陷，我们开发了一种无监督组件分割技术，该技术利用基础模型为轻量化分割网络自主生成训练标签，无需人工标注。通过将这一新型分割技术与我们提出的补丁直方图模块及局部-全局师生（LGST）模块相结合，我们在MVTec LOCO AD数据集中实现了95.3%的检测AUROC，超越了现有最优方法。此外，与多数现有方法相比，我们提出的方法具有更低的延迟与更高的吞吐量。