MLG-STPM:在标签噪音下进行强大的工业异常检测的元学习引导的STPM
Yu-Hang Huang1, Sio-Long Lo1, Zhen-Qiang Chen1
1Faculty of Innovation Engineering, Macau University of Science and Technology, Macau 999078, China.
Sensors (Basel, Switzerland)
|October 16, 2025
概括
这项研究引入了Meta-Learning Guided STPM (MLG-STPM),以改善工业图像异常检测 (IAD),尽管有噪音标签. 新的框架通过使用Evolving Meta-Set (EMS) 来提高稳定性,以实现更准确的质量控制.
科学领域:
- 计算机视觉 计算机视觉
- 机器学习 机器学习
- 人工智能的人工智能
背景情况:
- 工业图像异常检测 (IAD) 对于质量控制至关重要.
- 训练数据中的标签噪声显著降低了IAD的性能.
- 无监督的方法,特别是学生-教师框架,是克服标签噪音挑战的关键.
研究的目的:
- 开发一个新的框架,Meta-Learning Guided STPM (MLG-STPM),增强学生-教师特征金字塔匹配 (STPM) 对标签噪声的稳定性.
- 在没有外部清洁数据集或复杂的重权计划的情况下,减轻IAD中噪音标签的影响.
主要方法:
- 引入了MLG-STPM,这是一个由STPM和meta-learning启发的新框架.
- 整合了一个Evolving Meta-Set (EMS) 来动态地保持训练样本的高可信度子集.
- 训练了学生网络,使用当前批量和EMS的组合来减少噪音标签的影响.
主要成果:
- 与最初的STPM相比,MLG-STPM在异常检测和定位性能方面取得了显著的改进.
- 在更高的合成标签噪声条件下 (0%至20%),性能增长尤其显著.
- 与其他最先进的无监督异常检测方法相比,取得了具有竞争力的结果.
结论:
- 在面对标签噪声时,MLG-STPM有效地提高了学生-教师框架在工业图像异常检测中的稳定性.
- 拟议的Evolving Meta-Set机制为改善IAD准确性提供了一个可行的解决方案,而不需要依赖清洁的数据集.
相关概念视频
01:19Survival Tree
383
Survival trees are a non-parametric method used in survival analysis to model the relationship between a set of covariates and the time until an event of interest occurs, often referred to as the "time-to-event" or "survival time." This method is particularly useful when dealing with censored data, where the event has not occurred for some individuals by the end of the study period, or when the exact time of the event is unknown.
Building a Survival Tree
Constructing a...
Building a Survival Tree
Constructing a...
383
02:16Stereotype Content Model
15.3K
The Stereotype Content Model (SCM) was first proposed by Susan Fiske and her colleagues (Fiske, Cuddy, Glick & Xu, 2002; see also Fiske, 2012 and Fiske, 2017). The SCM specifies that when someone encounters a new group, they will stereotype them based on two metrics: warmth—or that group’s perceived intent, and how likely they are to provide help or inflict harm—and competence—or their ability to carry out that objective. Depending on the warmth-competence...
15.3K
01:02Quantifying and Rejecting Outliers: The Grubbs Test
3.5K
Sometimes, a data set can have a recorded numerical observation that greatly deviates from the rest of the data. Assuming that the data is normally distributed, a statistical method called the Grubbs test can be used to determine whether the observation is truly an outlier. To perform a two-tailed Grubbs test, first, calculate the absolute difference between the outlier and the mean. Then, calculate the ratio between this difference and the standard deviation of the sample. This...
3.5K