Search papers, labs, and topics across Lattice.
This paper introduces the Fuzzy Dual-dimensional Uncertainty (FuDU) framework for enhancing real-time industrial defect detection through a novel streaming active learning method. By employing a Prototype-based Global Uncertainty Quantification (PGUQ) module and a Dual-entropy defect Uncertainty Evaluator (DeUE), the approach effectively evaluates both image-level and box-level uncertainties, allowing for adaptive sampling decisions driven by expert knowledge. Experimental results confirm that FuDU significantly improves the reliability and efficiency of defect detection systems, particularly in high-stakes environments like nuclear fuel rod inspections.
FuDU transforms uncertainty into a strategic asset, enabling real-time defect detection with unprecedented reliability in industrial applications.
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.
