☆ 4.8 Article

SSCT-Net: A Semisupervised Circular Teacher Network for Defect Detection With Limited Labeled Multiview MFL Samples

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS (2023)

Related references

Note: Only part of the references are listed.

Article Engineering, Electrical & Electronic

A MFL Mechanism-Based Self-Supervised Method for Defect Detection With Limited Labeled Samples

He Zhao et al.

Summary: This article proposes a self-supervised method (MMSM) based on the magnetic flux leakage (MFL) mechanism, which can significantly improve defect detection accuracy and universality with limited labeled samples. The MMSM has three main characteristics: a hierarchical compression method to enhance weak defect signal features, a self-supervised network to reduce dependence on labeled data, and a novel DSDCN and T-SOHEM head network to enhance defect detection accuracy and reduce noise interference. Experimental results demonstrate the effectiveness of the proposed method in establishing a defect detection model with limited labeled samples.

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT (2023)

Add to Collection

Article Automation & Control Systems

An Intelligent Defect Detection Approach Based on Cascade Attention Network Under Complex Magnetic Flux Leakage Signals

Jinhai Liu et al.

Summary: A cascaded attention feature fusion network (CAFF-Net) is proposed in this article to improve the accuracy of pipeline defect detection under complex MFL signals. The network uses a novel feature enhancement method, a multilevel attention feature extraction module, and a feature aggregation module to extract explicit and implicit features of complex defects more effectively. The proposed method achieves a detection accuracy of 92% on average, which is 6.4% higher than the best result of state-of-the-art methods.

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS (2023)

Add to Collection

Article Geochemistry & Geophysics

WRMatch: Improving FixMatch With Weighted Nuclear-Norm Regularization for Few-Shot Remote Sensing Scene Classification

Yunsheng Xiong et al.

Summary: In this article, a novel framework is proposed for few-shot learning in remote sensing scene classification, utilizing the improved FixMatch and the weighted nuclear-norm regularization (WNNR). The method regularizes the prediction matrix using the nuclear-norm and provides two weighting schemes to improve the regularization. Experimental results demonstrate the effectiveness of the proposed method on three benchmark datasets.

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING (2022)

Add to Collection

Article Automation & Control Systems

A Multisensor Cycle-Supervised Convolutional Neural Network for Anomaly Detection on Magnetic Flux Leakage Signals

Lin Jiang et al.

Summary: A method called multisensor cycle-supervised convolutional neural network (CsCNN) is proposed for unsupervised anomaly detection in magnetic flux leakage (MFL) multisensor signals. The CsCNN includes multiple CNNs with the same structure and a cycle-supervised part, allowing for the establishment of latent relationships between multisensor signals and the application of a dynamic threshold for anomaly detection. Experiments show that the proposed method is effective compared to state-of-the-art methods.

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS (2022)

Add to Collection

Article Automation & Control Systems

Development of a Physics-Informed Doubly Fed Cross-Residual Deep Neural Network for High-Precision Magnetic Flux Leakage Defect Size Estimation

Hongyu Sun et al.

Summary: In this study, a physics-informed doubly fed cross-residual network (DfedResNet) is proposed for MFL defect detection, integrating physics-based MFL defect quantification theory into the neural network training. DfedResNet quantifies defects with high precision, especially in defect depth, by considering data from all three dimensions during training and using original magnetic signal data instead of recognized images. The DfedResNet model significantly reduces defect quantification errors and achieves a high quantification performance compared to other network structures and traditional algorithms.

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS (2022)

Add to Collection

Article Engineering, Electrical & Electronic

THMS-Net: A Two-Stage Heterogeneous Signals Mutual Supervision Network for MFL Weak Defect Detection

Lin Jiang et al.

Summary: A method based on a two-stage heterogeneous signals mutual supervision network is proposed for magnetic flux leakage (MFL) weak defect detection, improving the discrimination between defect and nondefect signals by enhancing their mutual supervision.

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

Rectifying Pseudo Label Learning via Uncertainty Estimation for Domain Adaptive Semantic Segmentation

Zhedong Zheng et al.

Summary: This paper proposes an approach to rectify pseudo label learning in unsupervised semantic segmentation adaptation by explicitly estimating prediction uncertainty during training and incorporating uncertainty into the optimization objective. Through extensive experiments, the proposed method achieves significant improvements on three benchmarks.

INTERNATIONAL JOURNAL OF COMPUTER VISION (2021)

Add to Collection

Article Instruments & Instrumentation

Defect identification based on wavelet decomposition for MFL non-destructive inspection of steel plates

D. Slesarev

Summary: The article discusses the use of wavelet decomposition for defect identification in magnetic flux leakage non-destructive testing, including criteria for choosing mother wavelet and a decision tree for defect classification. The algorithm has been tested on both simulated defects and real defects on an oil tank floor plate.

INSIGHT (2021)

Add to Collection

Article Computer Science, Information Systems

Weighted Pseudo Labeled Data and Mutual Learning for Semi-Supervised Classification

Jianwen Mo et al.

Summary: A semi-supervised classification algorithm involving weighted pseudo labeled data and mutual learning is proposed to improve classification performance and rectify incorrect pseudo labels. By utilizing selection, weighting, and mutual learning strategies, the accuracy of pseudo label predictions is effectively improved.

IEEE ACCESS (2021)

Add to Collection

Article Automation & Control Systems

Pipeline Magnetic Flux Leakage Image Detection Algorithm Based on Multiscale SSD Network

Lijian Yang et al.

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS (2020)

Add to Collection

Article Computer Science, Artificial Intelligence

A Semisupervised Recurrent Convolutional Attention Model for Human Activity Recognition

Kaixuan Chen et al.

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (2020)

Add to Collection

Article Engineering, Electrical & Electronic

Anomaly Detection of Complex MFL Measurements Using Low-Rank Recovery in Pipeline Transportation Inspection

Mingrui Fu et al.

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT (2020)

Add to Collection

Article Computer Science, Software Engineering

Software defect prediction using K-PCA and various kernel-based extreme learning machine: an empirical study

Sushant Kumar Pandey et al.

IET SOFTWARE (2020)

Add to Collection

Article Automation & Control Systems

An Estimation Method of Defect Size From MFL Image Using Visual Transformation Convolutional Neural Network

Senxiang Lu et al.

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS (2019)

Add to Collection

Article Computer Science, Information Systems

Adaptive Semi-Supervised Feature Selection for Cross-Modal Retrieval

En Yu et al.

IEEE TRANSACTIONS ON MULTIMEDIA (2019)

Add to Collection

Article Automation & Control Systems

A Highly Accurate Framework for Self-Labeled Semisupervised Classification in Industrial Applications

Di Wu et al.

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS (2018)

Add to Collection

Article Automation & Control Systems

An Adaptive Semisupervised Feature Analysis for Video Semantic Recognition

Minnan Luo et al.

IEEE TRANSACTIONS ON CYBERNETICS (2018)

Add to Collection

Article Computer Science, Artificial Intelligence

Semisupervised Feature Analysis by Mining Correlations Among Multiple Tasks

Xiaojun Chang et al.

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (2017)

Add to Collection

Article Engineering, Electrical & Electronic

Injurious or Noninjurious Defect Identification From MFL Images in Pipeline Inspection Using Convolutional Neural Network

Jian Feng et al.

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT (2017)

Add to Collection

Article Engineering, Civil

A clustering approach for assessing external corrosion in a buried pipeline based on hidden Markov random field model

Hui Wang et al.

STRUCTURAL SAFETY (2015)

Add to Collection

© Peeref 2019-2024. All rights reserved.