发表机构
Los Alamos National Laboratory; Triad National Security, LLC（洛斯阿拉莫斯国家实验室; Triad国家安全有限责任公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究通过评估19种无监督异常检测模型在BowTie制造数据集上的表现，揭示了基准性能与工业部署的差距，开发并部署了人机协同的制造部件检测框架。
AI 中文摘要
自动化异常检测方法在精心设计的学术基准上通常表现出色，但在真实工业场景中的表现尚不清楚。本研究在BowTie数据集上评估了19种无监督异常检测模型，该数据集是具有反光表面、细微缺陷和特定轮廓变化的挑战性制造数据集。与基准结果不同，研究发现模型性能比MVTec AD等标准基准上通常报告的更不稳定，对预处理高度敏感，且在不同条件下不一致，没有单一方法表现出普遍的鲁棒性；共识审计进一步表明，名义数据质量会影响部署。受这些发现的启发，我们开发并初步部署了一个用于制造部件检测的统一人机协同框架，该框架结合了图像标注、AI辅助缺陷检测和集成验证引擎，取代了之前的手动视觉检查和文档工作流程。该系统支持热力图引导的缺陷审查、SAM优化的候选区域供检查员接受、拒绝或边界调整、存在标注时的掩码评估，以及用于检查员一致性和入职培训的审查历史。这些结果凸显了基准性能与部署现实之间的差距，并提供了应对该差距的实用框架。
英文摘要
Automated anomaly detection methods often report strong performance on curated academic benchmarks, but their behavior under real-world industrial conditions is less clear. In this work, we evaluate 19 unsupervised anomaly detection models on the BowTie dataset, a challenging manufacturing dataset with reflective surfaces, subtle defects, and profile-specific variation. In contrast to benchmark results, we observe that model performance is less stable than typically reported on standard benchmarks such as MVTec AD, highly sensitive to preprocessing, and inconsistent across conditions, with no single approach emerging as uniformly robust; a consensus audit further indicates that nominal-data quality affects deployment. Motivated by these findings, we developed and initially deployed a unified human-in-the-loop framework for manufactured-part inspection that combines image annotation, AI-assisted defect detection, and an integrated validation engine, replacing a prior manual visual inspection and documentation workflow. The system supports heatmap-guided defect review, SAM-refined candidate regions for inspector acceptance, rejection, or boundary adjustment, mask evaluation where annotations exist, and review history for inspector consistency and onboarding. Together, the results highlight the gap between benchmark performance and deployment reality, and provide a practical framework for addressing it.
Comments8 pages, 8 figures, 6 tables. Accepted as a regular paper at the 25th International Conference on Machine Learning and Applications (ICMLA 2026)
