mlsanityData to 14 Oct 2026

FuDU: A Fuzzy Dual-dimensional Uncertainty Framework for Streaming Active Learning in Industrial Defect Detection

arXiv 2609.02212 ↗First posted 2 Sep 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Ensuring the reliability of deep learning models in real-time industrial defect detection is critical for high-stakes quality inspection. To mine uncertain samples within continuous industrial media streams, thereby enhancing the reliability of the detection system, this paper proposes a streaming active learning method based on the Fuzzy Dual-dimensional Uncertainty (FuDU) framework. Specifically, we first design a Prototype-based Global Uncertainty Quantification (PGUQ) module on the backbone to evaluate image-level uncertainty via normal/defective feature prototypes. A Dual-entropy defect Uncertainty Evaluator (DeUE) is then integrated into the detection head to quantify box-level uncertainty. Finally, by modeling uncertainty as systematic error, we propose a fuzzy dual-dimensional uncertainty-aware strategy that leverages fuzzy inference to fuse dual-dimensional uncertainties, enabling expert knowledge-driven adaptive sampling decisions. Comprehensive experiments demonstrate that FuDU is efficient and flexible, making it well-suited for challenging industrial inspection tasks such as the detection of nuclear fuel rod defects. Our code is publicly available at: https://github.com/wangzhaoyang-508/FuDU.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 25 of its 54 references that are in the map. First posted on arXiv 2 Sep 2026.

Builds on

25 of 54 references.

References · 25

  1. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  2. 2017Deep Bayesian Active Learning with Image Datacited by 819
  3. 2022Exploring Plain Vision Transformer Backbones for Object Detectioncited by 724
  4. 2023DETRs Beat YOLOs on Real-time Object Detectioncited by 533
  5. 2024AdaCLIP: Adapting CLIP with Hybrid Learnable Prompts for Zero-Shot Anomaly Detectioncited by 127
  6. 2021Active Learning for Deep Object Detection via Probabilistic Modelingcited by 79
  7. 2024A Unified Anomaly Synthesis Strategy with Gradient Ascent for Industrial Anomaly Detection and Localizationcited by 74
  8. 2024DEIM: DETR with Improved Matching for Fast Convergencecited by 68
  9. 2024A Survey on Deep Active Learning: Recent Advances and New Frontierscited by 54
  10. 2025Towards Zero-Shot Anomaly Detection and Reasoning with Multimodal Large Language Modelscited by 54
  11. 2019Deep Active Learning for Efficient Training of a LiDAR 3D Object Detectorcited by 45
  12. 2024Cross-Domain Few-Shot Object Detection via Enhanced Open-Set Object Detectorcited by 43
13 more
  1. 2020Towards Robust Pattern Recognition: A Reviewcited by 31
  2. 2023Active learning for data streams: a surveycited by 29
  3. 2022Plug and Play Active Learning for Object Detectioncited by 28
  4. 2023An Incremental Unified Framework for Small Defect Inspectioncited by 25
  5. 2024Coreset Selection for Object Detectioncited by 18
  6. 2022Active Learning Strategies for Weakly-supervised Object Detectioncited by 17
  7. 2023Defect Spectrum: A Granular Look of Large-Scale Defect Datasets with Rich Semanticscited by 16
  8. 2025MI-DETR: An Object Detection Model with Multi-time Inquiries Mechanismcited by 9
  9. 2024Few-shot Defect Image Generation based on Consistency Modelingcited by 8
  10. 2025Towards Real Unsupervised Anomaly Detection Via Confident Meta-Learningcited by 8
  11. 2025Uncertainty Meets Diversity: A Comprehensive Active Learning Framework for Indoor 3D Object Detectioncited by 7
  12. 2026MAU-GPT: Enhancing Multi-type Industrial Anomaly Understanding via Anomaly-aware and Generalist Experts Adaptationcited by 6
  13. 2023Classification Committee for Active Deep Object Detectioncited by 1

Built on by

Not cited yet.