paper-with-me

Papers

Composite Score for Anomaly Detection in Imbalanced Real-World Industrial Dataset

2022-11-25 · Arnaud Bougaham, Mohammed El Adoui, Isabelle Linden, Benoît Frénay

In recent years, the industrial sector has evolved towards its fourth revolution. The quality control domain is particularly interested in advanced machine learning for computer vision anomaly detection. Nevertheless, several challenges have to be faced, including imbalanced datasets, the image complexity, and the zero-false-negative (ZFN) constraint to guarantee the high-quality requirement. This paper illustrates a use case for an industrial partner, where Printed Circuit Board Assembly (PCBA) images are first reconstructed with a Vector Quantized Generative Adversarial Network (VQGAN) trained on normal products. Then, several multi-level metrics are extracted on a few normal and abnormal images, highlighting anomalies through reconstruction differences. Finally, a classifer is trained to build a composite anomaly score thanks to the metrics extracted. This three-step approach is performed on the public MVTec-AD datasets and on the partner PCBA dataset, where it achieves a regular accuracy of 95.69% and 87.93% under the ZFN constraint.

📄 PDF Abstract BibTeX arXiv:2211.15513

Code (0)

등록된 구현이 없습니다.

Tasks

Anomaly DetectionGenerative Adversarial Network

Similar Papers 제목 키워드 기반

Fair Anomaly Detection For Imbalanced Groups

2024-09-17 · Ziwei Wu, Lecheng Zheng, Yuancheng Yu, Ruizhong Qiu 외

Anomaly detection (AD) has been widely studied for decades in many real-world applications, including fraud detection in finance, and intrusion detection for cybersecurity, etc. Due to the imbalanced nature between prote…

Anomaly DetectionContrastive LearningFairnessFraud Detection+1

Adversarial Anomaly Detection using Gaussian Priors and Nonlinear Anomaly Scores

2023-10-27 · Fiete Lüer, Tobias Weber, Maxim Dolgich, Christian Böhm

Anomaly detection in imbalanced datasets is a frequent and crucial problem, especially in the medical domain where retrieving and labeling irregularities is often expensive. By combining the generative stability of a $\b…

Anomaly Detection

Multi-Cue Anomaly Detection and Localization under Data Contamination

2026-01-30 · Anindya Sundar Das, Monowar Bhuyan arxiv

Visual anomaly detection in real-world industrial settings faces two major limitations. First, most existing methods are trained on purely normal data or on unlabeled datasets assumed to be predominantly normal, presumin…

Anomaly Detection

Robust Anomaly Detection with Graph Neural Networks using Controllability

2025-07-18 · Yifan Wei, Anwar Said, Waseem Abbas, Xenofon Koutsoukos arxiv

Anomaly detection in complex domains poses significant challenges due to the need for extensive labeled data and the inherently imbalanced nature of anomalous versus benign samples. Graph-based machine learning models ha…

Anomaly Detection

Towards Real-World Rumor Detection: Anomaly Detection Framework with Graph Supervised Contrastive Learning

2025-08-10 · Chaoqun Cui, Caiyan Jia arxiv

Current rumor detection methods based on propagation structure learning predominately treat rumor detection as a class-balanced classification task on limited labeled data. However, real-world social media data exhibits …

Contrastive LearningAnomaly Detection