相关参考文献
注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article
Engineering, Multidisciplinary
Qi-Ang Wang et al.
Summary: Central to structural health monitoring is the modeling, manipulation, and interpretation of data using a sophisticated system. A novel application of a fully probabilistic and high-precision data modeling method, called improved Sparse Bayesian Learning, was proposed to address the challenge of accurately modeling and forecasting SHM measurements. The proposed method automates sparsity of the Bayesian model based on the features of SHM monitoring data, improving the generalization ability and data prediction accuracy. Embedded in a Bayesian framework with built-in protection against over-fitting, the proposed scheme exhibits high robustness to data noise, especially for data forecasting, and has been verified on real-world vibration data collected from a large-scale cable-stayed bridge.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2023)
Article
Engineering, Electrical & Electronic
Hanwei Zhao et al.
Summary: This article proposes a bridge deflection modeling strategy based on Bayesian multiple linear regression, which effectively deals with the interference of data time lag and abnormal signals, achieves dynamic updating capability, and outperforms traditional methods and deep learning methods in terms of accuracy, storage space, and time. In the case of sparse sensing points and insufficient monitoring time, the deflection of newly added or adjusted sensing points can be quickly predicted using short-term observation data.
IEEE SENSORS JOURNAL
(2023)
Article
Engineering, Multidisciplinary
Qi Li et al.
Summary: Outlier detection is a crucial procedure in structural health monitoring to ensure clean and reliable data. This study proposes a robust time series outlier detection method that combines a Bayesian perspective and an extreme learning machine neural network model. The proposed method effectively detects outliers by quantifying their probability and shows improved model prediction performance after outlier removal.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2023)
Article
Engineering, Multidisciplinary
Haoyu Zhang et al.
Summary: In the field of structural health monitoring (SHM), real-time image-based damage detection is highly demanded. Compressive sampling (CS), a promising solution to reduce power consumption, has faced limitations such as low compression ratios and the need for laborious training. To overcome these barriers, a multitask CS algorithm that relies on existing generators trained by low-pixel crack images is proposed. By exploiting the similarity in sparsity pattern among crack images, the algorithm achieves higher crack detection accuracy and efficiency. Verification using synthetic and real image data shows its potential for operational CS-based crack detection systems in real-time SHM.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2023)
Article
Engineering, Civil
Qi-Ang Wang et al.
Summary: The study utilized the VHGP model for modeling and forecasting strain data during typhoon events in structural health monitoring, showing that the VHGP model has better regression accuracy and robustness, can evaluate noise level changes, and provides a basis for structural health assessment of large bridges.
ENGINEERING STRUCTURES
(2022)
Article
Engineering, Mechanical
Xiuli Chai et al.
Summary: This paper proposes an effective visually meaningful color image encryption scheme by combining HMPSO, BCS and HD algorithms. The scheme uses techniques such as discrete cosine transform, chaotic sequence generation, and embedding algorithm to enhance the security and efficiency of image encryption. By optimizing the threshold value of sparse coefficient modification and the embedding rate, the quality of the reconstructed image and cipher image is improved.
NONLINEAR DYNAMICS
(2022)
Article
Engineering, Multidisciplinary
Qi-Ang Wang et al.
Summary: A Bayesian dynamic linear model (BDLM) framework is proposed for data modeling and forecasting to evaluate the performance of Ting Kau Bridge in Hong Kong under extreme typhoon loads. The study shows that the BDLM framework yields more accurate calculations compared to Gaussian process and Variational Heteroscedasticity Gaussian Process methods for one-step ahead forecasting of strain data under typhoons. Multi-step ahead forecasting is also successfully carried out within an acceptable precision range for both non-typhoon and typhoon responses.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2022)
Article
Construction & Building Technology
Qi-Ang Wang et al.
Summary: Despite the continuous evolution and development of structural health monitoring (SHM) technology, it is still a challenge to interpret a large amount of sensed data and extract useful information about the structural health condition. To resolve this problem, a novel application of the probabilistic data-driven damage detection method based on Sparse Bayesian Learning (SBL) scheme was proposed. This method constructs a new structural damage index using vibration monitoring data and establishes an SBL regression model as a reference base. The damage detection and discrimination capabilities of this methodology were verified using field monitoring data from a cable-stayed bridge.
STRUCTURAL CONTROL & HEALTH MONITORING
(2022)
Article
Engineering, Electrical & Electronic
Zhifeng Wang et al.
Summary: This paper proposes a novel framework based on deep learning for compressed sensing image measurement and reconstruction. By avoiding the block effect and utilizing multi-scale feature extraction, this method outperforms existing methods in terms of image reconstruction performance.
CIRCUITS SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Computer Science, Artificial Intelligence
Bin Chen et al.
Summary: CASNet is a novel content-aware scalable network that efficiently addresses image compressed sensing (CS) problems by achieving adaptive sampling rate allocation and high-quality reconstruction. It utilizes a data-driven saliency detector to evaluate the importance of different image regions and a saliency-based strategy for sampling rate allocation. By using a unified learnable generating matrix and an optimization-inspired recovery subnet, CASNet jointly reconstructs image blocks sampled at various sampling rates. The proposed SVD-based initialization and random transformation enhancement strategies improve training convergence and network robustness. Experiments show that CASNet outperforms other CS networks, demonstrating the collaboration and mutual supports among its components and strategies.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2022)
Article
Engineering, Multidisciplinary
Xiaoming Lei et al.
Summary: This study introduces a deep convolutional generative adversarial network for reconstructing lost data in structural health monitoring. The generator is trained to extract features from the data set and reconstruct lost signals using responses from remaining functional sensors, while the discriminator provides feedback to improve reconstruction accuracy. Training the model involves using reconstruction loss and adversarial loss to handle low-frequency and high-frequency features of the signals, demonstrating effectiveness through case studies.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2021)
Review
Engineering, Civil
Sandeep Sony et al.
Summary: With recent advancements in non-contact sensing technology, there has been a significant growth in deep learning-based condition assessment techniques for structural systems, primarily relying on convolutional neural networks (CNNs). Researchers have developed various CNN architectures to adapt to different conditions and improve the detection and analysis of structural damage. These techniques have been applied in infrastructure monitoring and maintenance.
ENGINEERING STRUCTURES
(2021)
Article
Engineering, Mechanical
Yong Huang et al.
Summary: This paper introduces a new CS method that replaces sparsity regularization with a generative model for automatic crack segmentation of compressed crack images. The method utilizes generative models to capture necessary features, demonstrating remarkable performance and energy savings.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2021)
Article
Construction & Building Technology
Mehmet Ergun Hatir et al.
Summary: The petrographic determination and mapping model of building stone, developed based on Mask R-CNN, can provide important bases for restoration works with its fast and accurate mapping capability.
JOURNAL OF BUILDING ENGINEERING
(2021)
Article
Archaeology
Ergun Hatir et al.
Summary: This study developed a model using the Mask R-CNN algorithm to automatically detect and map deterioration in archaeological heritage sites. Trained on 2460 images of Yazilikaya monuments in the Hattusa archaeological site, the model was able to successfully distinguish deterioration from other anomalies and apply successfully in monuments in different regions.
JOURNAL OF CULTURAL HERITAGE
(2021)
Article
Computer Science, Interdisciplinary Applications
FuTao Ni et al.
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING
(2020)
Article
Chemistry, Analytical
Tongwei Liu et al.
Article
Computer Science, Artificial Intelligence
Hantao Yao et al.
Article
Construction & Building Technology
Gao Fan et al.
STRUCTURAL CONTROL & HEALTH MONITORING
(2019)
Article
Multidisciplinary Sciences
Andrius Vabalas et al.
Article
Engineering, Electrical & Electronic
Michael Iliadis et al.
DIGITAL SIGNAL PROCESSING
(2018)
Article
Computer Science, Interdisciplinary Applications
Guang Yang et al.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2018)
Article
Geochemistry & Geophysics
Su Liu et al.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2017)
Article
Engineering, Electrical & Electronic
Hamid Palangi et al.
IEEE TRANSACTIONS ON SIGNAL PROCESSING
(2016)
Article
Engineering, Mechanical
Yong Huang et al.
PROBABILISTIC ENGINEERING MECHANICS
(2016)
Article
Computer Science, Interdisciplinary Applications
Yong Huang et al.
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING
(2014)
Article
Engineering, Electrical & Electronic
Fei Zhou et al.
EURASIP JOURNAL ON ADVANCES IN SIGNAL PROCESSING
(2014)
Article
Mathematics, Applied
D. Needell et al.
APPLIED AND COMPUTATIONAL HARMONIC ANALYSIS
(2009)
Article
Engineering, Electrical & Electronic
Shihao Ji et al.
IEEE TRANSACTIONS ON SIGNAL PROCESSING
(2008)
Article
Engineering, Electrical & Electronic
Mario A. T. Figueiredo et al.
IEEE JOURNAL OF SELECTED TOPICS IN SIGNAL PROCESSING
(2007)
Article
Computer Science, Information Systems
DL Donoho
IEEE TRANSACTIONS ON INFORMATION THEORY
(2006)
Article
Computer Science, Information Systems
EJ Candès et al.
IEEE TRANSACTIONS ON INFORMATION THEORY
(2006)