发表机构
FPT University（越南胡志明市邮电大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对基于重建的图像异常检测易受异常值泄漏影响的问题，提出非线性重建损失及统计校准方案，通过抑制高幅值特征和数据驱动控制抑制强度，提升异常检测性能，在多个数据集上取得良好结果。
AI 中文摘要
基于重建的方法是无监督图像异常检测的基石，但易受“异常值泄漏”影响，标准均方误差（MSE）损失会使模型忠实地重建异常模式。我们提出一种非线性重建损失，应用基于sigmoid的挤压函数抑制高幅值特征，防止异常值主导优化，同时保持对正常模式的敏感性。此外，引入统计校准方案，从正常特征分布的置信区间选择缩放因子k，实现数据驱动的抑制强度控制。与现有方法相比，我们的方法在异常检测性能上具有竞争力或更优。例如在MVTec-AD上达到99.0%的图像AUROC和97.3%的像素AUROC，在VisA上达到95.3%的图像AUROC和99.0%的像素AUROC。这些结果表明非线性梯度抑制是减轻异常值泄漏和改善统一工业检测设置中异常定位的有效机制。
英文摘要
Reconstruction-based methods are a cornerstone of unsupervised image anomaly detection, but they remain vulnerable to \emph{outlier leakage}, where standard mean squared error (MSE) loss drives the model to faithfully reconstruct anomalous patterns. We propose a Non-linear Reconstruction Loss that applies a sigmoid-based squashing function to suppress high-magnitude features, preventing outliers from dominating optimization while preserving sensitivity to normal patterns. In addition, we introduce a statistical calibration scheme that selects the scaling factor $k$ from the confidence interval (CI) of the normal feature distribution, enabling data-driven control of the suppression strength. Our approach achieves competitive or superior anomaly detection performance compared to state-of-the-art methods, reaching 99.0\% Image-AUROC and 97.3\% Pixel-AUROC on MVTec-AD, and 95.3\% Image-AUROC and 99.0\% Pixel-AUROC on VisA. These results indicate that non-linear gradient suppression is an effective mechanism for mitigating outlier leakage and improving anomaly localization in unified industrial inspection settings. The implementation is available at https://github.com/mintii13/Statistical-Non-linear-Reconstruction-Loss.git.
CommentsAccepted at KES 2026
