Dual-Metric Neural Network With Attention Guidance for Surface Defect Few-Shot Detection in Smart Manufacturing
- 17 October 2023
- journal article
- research article
- Published by ASME International in Journal of Manufacturing Science and Engineering
- Vol. 145 (12) , 1-31
Abstract
As an important application of human–robot collaboration, intelligent detection of surface defects is crucial for production quality control, which also helps in relieving the workload of technical staff in human-centric smart manufacturing. To accurately detect defects with limited samples in industrial practice, a dual-metric neural network with attention-guided is proposed. First, an attention-guided recognition network with channel attention and position attention module is designed to efficiently learn representative defect features with limited samples. Second, aiming to detect defects with confusing surface images, a dual-metric function is presented to learn the classification boundary by controlling the distance of samples in feature space from intraclass and interclass. The experiment results on the fabric defect dataset demonstrate that the proposed approach outperforms state-of-the-art methods in accuracy, recall, precision, F1-score, and few-shot accuracy. Further comparative experiments reveal that the dual-metric function is superior in improving the few-shot detection accuracy.Keywords
Funding Information
- CAST Innovation Foundation (2021QNRC001)
- Donghua University (GSIF-DH-M-2022007)
- National Natural Science Foundation of China (52275478)
- Natural Science Foundation of Shanghai Municipality (22ZR1403000)
- Natural Science Foundation of Xinjiang Uygur Autonomous Region (2022B01057-1)
This publication has 29 references indexed in Scilit:
- A semi-supervised convolutional neural network-based method for steel surface defect recognitionRobotics and Computer-Integrated Manufacturing, 2020
- A Multilevel Information Fusion-Based Deep Learning Method for Vision-Based Defect RecognitionIEEE Transactions on Instrumentation and Measurement, 2019
- Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based LocalizationInternational Journal of Computer Vision, 2019
- Design of multi-scale receptive field convolutional neural network for surface inspection of hot rolled steelsImage and Vision Computing, 2019
- AdaBalGAN: An Improved Generative Adversarial Network With Imbalanced Learning for Wafer Defective Pattern RecognitionIEEE Transactions on Semiconductor Manufacturing, 2019
- A Generic Deep-Learning-Based Approach for Automated Surface InspectionIEEE Transactions on Cybernetics, 2017
- Deep Feature Extraction and Classification of Hyperspectral Images Based on Convolutional Neural NetworksIEEE Transactions on Geoscience and Remote Sensing, 2016
- An Overview and Empirical Comparison of Distance Metric Learning MethodsIEEE Transactions on Cybernetics, 2016
- Influence of cutting environments on surface integrity and power consumption of austenitic stainless steelRobotics and Computer-Integrated Manufacturing, 2015
- Analysis of Multinomial Models With Unknown Index Using Data AugmentationJournal of Computational and Graphical Statistics, 2007