Article
Geochemistry & Geophysics
Yinghui Quan, Rui Zhang, Yachao Li, Ran Xu, Shengqi Zhu, Mengdao Xing
Summary: This article proposes a new microwave forward-looking correlated 3-D imaging method based on random radiation field combined with sparse reconstruction, which can break through Rayleigh resolution limitation and achieve resolution at least 5.5 times higher than real aperture imaging. Additionally, an improved quasi-Newton iteration method based on GPU platform is developed to improve computation efficiency of sparse reconstruction.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Chemistry, Analytical
Miao Xu, Chao Wang, Kaikai Wang, Haodong Shi, Yingchao Li, Huilin Jiang
Summary: This paper proposes a novel polarization imaging method based on deep compressed sensing, which reconstructs high-resolution images under low sampling rate conditions by adding digital micromirror devices to an optical system and simulating the polarization transmission model. The feasibility of our approach is validated by constructing a simulated dataset, training a polarization super-resolution imaging network, and demonstrating excellent reconstructions on real shooting scenes.
Article
Engineering, Electrical & Electronic
Jingwei He, Lei Yu, Zhou Liu, Wen Yang
Summary: This paper introduces an end-to-end trainable super-resolution model that leverages both deep learning and prior-based methods, using reweighted algorithm and learning weighted iterative soft thresholding algorithm in a convolutional manner. Extensive experiments demonstrate the superiority of this method over recent SISR methods in terms of both quantitative and qualitative results.
SIGNAL IMAGE AND VIDEO PROCESSING
(2021)
Article
Computer Science, Artificial Intelligence
Peilin Chen, Wenhan Yang, Meng Wang, Long Sun, Kangkang Hu, Shiqi Wang
Summary: This paper proposes a novel approach for deep video super-resolution in the compressed domain, utilizing coding priors and deep priors to reconstruct high-resolution videos effectively. The incorporation of the GSFT layer and guided soft alignment scheme, combining spatial and temporal coding priors, leads to more effective video reconstruction.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Computer Science, Information Systems
Meng Wang, Jizheng Xu, Li Zhang, Junru Li, Kai Zhang, Shiqi Wang, Siwei Ma
Summary: This article focuses on the super-resolution problem of screen content images and proposes a dataset for handling screen contents with different levels of compression distortion. The principle of multi-hypothesis is introduced into super-resolution, and a new paradigm is proposed for restoring compressed screen content images. The luminance and sharpness similarity metric is adopted in network learning to better adapt to the screen content characteristic and ensure perceptual fidelity. Experimental results verify the superiority and effectiveness of the proposed method.
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
(2023)
Article
Operations Research & Management Science
Shuqin Sun, Ting Kei Pong
Summary: We propose a new algorithmic framework for constrained compressed sensing models with nonconvex sparsity-inducing regularizers and nonconvex loss functions. Our framework efficiently solves subproblems using well-developed solvers and introduces a new termination criterion for infeasible solutions. Numerical comparisons demonstrate that our approaches outperform existing methods in terms of recovery errors and speed, particularly for badly scaled measurement matrices.
COMPUTATIONAL OPTIMIZATION AND APPLICATIONS
(2023)
Article
Engineering, Aerospace
Min Wu, Chengpeng Hao
Summary: This article presents a compressed sensing-based algorithm for TOA and AOA estimation in OFDM radar systems, which optimizes the CP-OFDM signal structure to achieve effective target estimation and interference suppression.
IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS
(2022)
Article
Engineering, Mechanical
Shijia Yin, Li Li, Yang Zhao, Linyong Li, Yang Yang, Zhigang Chu
Summary: This paper proposes an off-grid sparse Bayesian inference-based compressive spherical beamforming (OGSBI-CSB) method, which effectively solves the basis mismatch problem in traditional CSB and significantly improves identification accuracy, super-resolution, and resistance to noise interference for off-grid sources.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Engineering, Electrical & Electronic
Alaa M. M. El-Ashkar, Taha El Sayed Taha, Adel S. S. El-Fishawy, Mohammed Abd-Elnaby, Fathi E. E. Abd El-Samie, Walid El-Shafai
Summary: This paper proposes the use of Compressed Sensing (CS) for SAR image compression and Single-Image Super-Resolution (SIMSR) for image enhancement to improve target detection performance.
OPTICAL AND QUANTUM ELECTRONICS
(2023)
Article
Optics
Wuyang Zhang, Ping Song, Xuanquan Wang, Zhaolin Zheng, Yunjian Bai, Haocheng Geng
Summary: This paper proposes a fast and lightweight compressive ToF framework for super-resolution imaging. By introducing the block compressed sensing method, the reconstruction time and data storage requirements are significantly decreased, providing a development direction for compressive ToF imaging.
Article
Computer Science, Information Systems
Yan Wang, Minggang Dong, Wei Ye, Deao Liu, Guojun Gan
Summary: This article proposes a contrastive learning-based iterative network (CLIN) for noisy remote sensing image super-resolution. CLIN addresses the issue of noise impact by utilizing an evaluator and a reconstructor to evaluate noise levels and reconstruct super-resolution images. Additionally, a global feature fusion block and a contrastive penalty strategy are introduced to enhance image quality and suppress noise.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Engineering, Electrical & Electronic
Xingyu Tuo, Yin Zhang, Yulin Huang, Jianyu Yang
Summary: A method for fast azimuth super-resolution imaging of RAR based on IRLS and LS has been proposed in this article, effectively addressing the azimuth resolution issue in radar forward-looking imaging, reducing time complexity without sacrificing imaging performance.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Article
Engineering, Biomedical
Jihun Kim, Qingfei Wang, Siyuan Zhang, Sangpil Yoon
Summary: The study developed an SRUS imaging technique based on L1H-CS which locates densely populated MBs to visualize microvasculature. Compared to traditional techniques, L1H-CS showed better performance in data acquisition time and image quality.
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING
(2021)
Article
Computer Science, Information Systems
Ilhwan Kwon, Jun Li, Mukesh Prasad
Summary: Video compression technology for UHD and 8K UHD video has been widely adopted by major broadcasting companies and video content providers. However, broadcasting high-resolution video content is still challenging due to limited network bandwidth and data storage. A solution is to downsample the video at the transmission side and utilize Video Super-Resolution (VSR) technology at the receiving end to recover the original quality.
Article
Geochemistry & Geophysics
Hayatomomaru Morimoto, Shouhei Kidera
Summary: This study presents a super-resolution subsurface imaging method based on sparse regularization for multilayer structure analysis in the terahertz band. The depth resolution dependence in optical lens-based measurements can be solved by using an appropriate depth-dependent reference signal.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Engineering, Electrical & Electronic
Ananth Narayan Samudrala, M. Hadi Amini, Soummya Kar, Rick S. Blum
IEEE TRANSACTIONS ON SMART GRID
(2020)
Article
Engineering, Electrical & Electronic
Anantha K. Karthik, Rick S. Blum
IEEE TRANSACTIONS ON COMMUNICATIONS
(2020)
Article
Engineering, Electrical & Electronic
Yongjun Liu, Rick S. Blum, Guisheng Liao, Shengqi Zhu
Article
Engineering, Industrial
Mingxing Li, Daqiang Guo, Ming Li, Ting Qu, George Q. Huang
Summary: The widespread adoption of Industry 4.0 technologies is revolutionising manufacturing operations. This paper introduces a novel concept of operations twins (OT) for achieving synchronisation between production and intralogistics (PiL) through the use of Industry 4.0 technologies and innovative operations management strategies.
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH
(2023)
Article
Engineering, Multidisciplinary
Mianjie Li, Yuan Liu, Zhihong Tian, Chun Shan
Summary: In this article, an information hiding method based on multidimensional feature fusion for privacy protection in 6G networks is proposed. The method is based on the strong attack resistance of carriers carrying private information, and it involves studying multidimensional feature fusion arbitration methods and constructing an efficient and accurate feature-based information hiding technology.
IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Abdulkadir Celikkanat, Yanning Shen, Fragkiskos D. Malliaros
Summary: This paper proposes a weighted matrix factorization model that incorporates random walk-based information for learning node representations in network analysis. The novel formulation allows the utilization of kernel functions to enhance the expressiveness of existing matrix decomposition methods and alleviate their computational complexities. An empirical evaluation on real-world networks demonstrates that the proposed model outperforms baseline node embedding algorithms in downstream machine learning tasks.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Proceedings Paper
Acoustics
Pouya M. Ghari, Yanning Shen
Summary: This paper presents an algorithmic framework for addressing communication bandwidth limitations in federated learning. By learning an ensemble of pre-trained models, the structure of the ensemble model is constructed based on the server's confidence in the models. Only selected models are transmitted to the clients to ensure certain budget constraints are not violated. Experimental results demonstrate the effectiveness of the proposed approach.
2022 30TH EUROPEAN SIGNAL PROCESSING CONFERENCE (EUSIPCO 2022)
(2022)
Proceedings Paper
Acoustics
Oyku Deniz Kose, Yanning Shen
Summary: This study modifies existing online learning and selective sampling algorithms to be used with graphs that have nodal features. A bias reduction strategy is proposed and experiments on real social networks demonstrate the advantages of incorporating nodal features and a fairness-enhancement framework.
2022 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP)
(2022)
Proceedings Paper
Acoustics
Pouya M. Ghari, Yanning Shen
Summary: Online learning with expert advice is a widely used approach in various machine learning tasks, considering the relationship among experts. This study presents a novel online learning algorithm to handle uncertainties in expert relationships and proves its effectiveness under certain conditions.
2022 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP)
(2022)
Article
Engineering, Electrical & Electronic
Oyku Deniz Kose, Yanning Shen
Summary: This study addresses the fairness issue in graph contrastive learning by proposing novel fairness-aware graph augmentations based on adaptive feature masking and edge deletion. By introducing different fairness notions to guide the design, it successfully reduces the intrinsic bias.
IEEE TRANSACTIONS ON SIGNAL AND INFORMATION PROCESSING OVER NETWORKS
(2022)
Article
Computer Science, Artificial Intelligence
Maosen Li, Siheng Chen, Yanning Shen, Genjia Liu, Ivor W. Tsang, Ya Zhang
Summary: This article focuses on predicting future statuses of multiple agents by utilizing dynamic interactions in the system. A novel collaborative prediction unit (CoPU) is proposed, which aggregates predictions from multiple collaborative predictors based on a collaborative graph. Experimental results demonstrate the superior performance of our method across different tasks.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Proceedings Paper
Acoustics
Zixiao Zong, Yanning Shen
Summary: The paper proposes a multi-kernel-based approach to reconstruct the attributes of unobserved nodes by adaptively combining the effects of multi-hop neighbors with global information, showing flexibility and the merit of the algorithm confirmed in experiments on real-world datasets.
2021 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP 2021)
(2021)
Article
Engineering, Electrical & Electronic
Yanning Shen, Saeed Karimi-Bidhendi, Hamid Jafarkhani
Summary: The paper proposes an online multi-kernel learning scheme to infer nonlinear functions from data samples collected at distributed locations; it studies the effects of quantization for communication efficiency among distributed nodes, and develops a distributed and quantized online multiple kernel learning algorithm; regret analysis shows the algorithm can achieve sublinear regret, and numerical tests on real datasets demonstrate its effectiveness.
IEEE TRANSACTIONS ON SIGNAL PROCESSING
(2021)
Article
Engineering, Electrical & Electronic
Ananth Narayan Samudrala, M. Hadi Amini, Soummya Kar, Rick S. Blum
IEEE TRANSACTIONS ON SMART GRID
(2020)
Article
Computer Science, Theory & Methods
Zisheng Wang, Rick S. Blum
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2020)