当更多参考样本反而有害：面向少样本钢铁缺陷检测的感知污染的DINOv2记忆库

中文摘要 AI 辅助

基于补丁记忆的异常检测器假设其参考库是正常的，而当存在额外未验证的工业图像时，这一假设难以保证。我们研究是否可以仅用少量可信的正常图像，在无缺陷掩码的情况下从这类参考样本中安全恢复有用的正常补丁。从AnomalyDINO所使用的DINOv2补丁记忆公式出发，我们通过候选补丁与干净种子库的距离对其评分，丢弃最可疑的20%，将保留的补丁与种子库合并，并通过贪心核心集选择来控制固定预算。在Severstal数据集上，单纯添加的额外参考样本包含9.46%的异常补丁；所提出的修剪方法可拒绝其中78.1%的异常补丁，将残留污染降至2.59%。在51200个补丁的相同开发预算下，所提出的记忆库达到0.1084的AUPRC，而单纯扩展的AUPRC为0.0950，随机移除的为0.0952，使用8张干净图像的为0.1030。仅向干净库中注入0.5%的异常补丁，就会使AUPRC从0.1030降至0.0759。在所有5个完成的保留对中，所提出的记忆库均优于单纯扩展，平均AUPRC提升0.0142。因此，参考纯度是一阶设计变量，未验证的图像仅在其贡献被显式过滤时才有用。

英文摘要

Patch-memory anomaly detectors assume that their reference bank is normal, an assumption that is difficult to guarantee when additional industrial images are unverified. We study whether a few trusted normal images can safely recover useful normal patches from such references without defect masks. Starting from the DINOv2 patch-memory formulation used by AnomalyDINO, we score candidate patches by distance to a clean seed bank, discard the most suspicious 20%, merge the retained patches with the seed, and enforce a fixed budget by greedy coreset selection. On Severstal, naive additional references contain 9.46% anomalous patches; the proposed trim rejects 78.1\% of them and reduces residual contamination to 2.59%. At an equal 51,200-patch development budget, the proposed bank reaches 0.1084 AUPRC versus 0.0950 for naive expansion, 0.0952 for random removal, and 0.1030 for eight clean images. Injecting only 0.5\% anomalous patches into a clean bank reduces AUPRC from 0.1030 to 0.0759. On all five completed held-out pairs, the proposed bank improves over naive expansion, with a mean gain of 0.0142 AUPRC. Reference purity is therefore a first-order design variable, and unverified images are useful only when their contribution is filtered explicitly.

发表机构

• University of Padua（帕多瓦大学）
• University of Vienna（维也纳大学）

机构由 AI 辅助整理，请以论文原文为准。