A5078574617- Robotics and Sensor-Based Localization
- Advanced Image and Video Retrieval Techniques
- Indoor and Outdoor Localization Technologies
- Fault Detection and Control Systems
- Target Tracking and Data Fusion in Sensor Networks
- Advanced Adaptive Filtering Techniques
- Advanced Vision and Imaging
- GNSS positioning and interference
- Advanced Control Systems Optimization
- Advanced Computational Techniques and Applications
- Image and Signal Denoising Methods
- Speech and Audio Processing
- Machine Fault Diagnosis Techniques
- Advanced Image Processing Techniques
- Blind Source Separation Techniques
- Advanced Neural Network Applications
- Underwater Vehicles and Communication Systems
- Advanced MRI Techniques and Applications
- Inertial Sensor and Navigation
- Infrared Target Detection Methodologies
- Energy Efficient Wireless Sensor Networks
- Cardiac Imaging and Diagnostics
- Advanced Image Fusion Techniques
- Spectroscopy and Chemometric Analyses
- Agricultural Economics and Policy
Taiyuan University of Technology
2016-2025
Hong Kong University of Science and Technology
2021-2024
University of Hong Kong
2021-2024
State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing
2023-2024
Union Hospital
2023-2024
Huazhong University of Science and Technology
2023-2024
Zhejiang International Studies University
2023
Zhejiang University
2023
Southwest University
2021-2022
State Key Laboratory of Silkworm Genomic Biology
2021-2022
Abstract The silkworm Bombyx mori is an important economic insect for producing silk, the “queen of fabrics”. currently available genomes limit understanding its genetic diversity and discovery valuable alleles breeding. Here, we deeply re-sequence 1,078 silkworms assemble long-read 545 representatives. We construct a high-resolution pan-genome dataset representing almost entire genomic content in silkworm. find that population harbors high density variants identify 7308 new genes, 4260...
In this paper, the dynamic CoVaR method is used to measure changes in systemic risk financial industry during COVID-19 pandemic. We find that, first, after outbreak of pandemic, increased significantly. Second, impact pandemic on securities was greater than that banking and insurance industries.
Automated classification of breast cancer subtypes from digital pathology images has been an extremely challenging task due to the complicated spatial patterns cells in tissue micro-environment. While newly proposed graph transformers are able capture more long-range dependencies enhance accuracy, they largely ignore topological connectivity between nodes, which is nevertheless critical extract representative features address this difficult task. In paper, we propose a novel...
Video Object Tracking (VOT) is a critical task in computer vision. While Siamese-based and Transformer-based trackers are widely used VOT, they struggle to perform well on the OTB100 benchmark due lack of dedicated training sets. This challenge highlights difficulty effectively generalizing unknown data. To address this issue, paper proposes an innovative method that utilizes tensor decomposition, underexplored concept object-tracking research. By applying L1-norm video sequences represented...
In this paper, a deep learning fault detection and prediction framework combining principal component analysis (PCA) Informer is proposed to solve the problem of online monitoring nuclear power valves which hard implement. More specifically, PCA plays role dimensionality reduction feature extraction. It maps data with multi-dimensional space low-dimensional extracts main features. At same time, T-square Q statistic thresholds are also provided realize abnormal status monitoring. Meanwhile,...
Unbalanced data with very few samples for special abnormal conditions frequently occur in actual production processes, which can make accurate monitoring of the process state challenging.This paper proposes a multi-modal few-shot learning method (MMFSL) within fault diagnosis framework unbalanced modelling industrial bearings.MMFSL handle two modes and therefore contains generation channels.The first channel deals time series second images.The generated are then evaluated using MMFSL...
For high-precision positioning and navigation systems, the precision of receiver is jeopardized by multipath interference. Multipath suppression methods based on data processing have drawn much attention recently. The critical step processing-based method to estimate parameters. However, most falling into category are limited Gaussian noises, which means performance these may be degraded in non-Gaussian noises encountered quite often reality. Besides, only static case studied existing...
Increasing the combustion efficiency of power plant boilers and reducing pollutant emissions are important for energy conservation environmental protection. The boiler process is a complex multi-input/multi-output system, with high degree nonlinearity strong coupling characteristics. It necessary to optimize model by means artificial intelligence methods. However, traditional intelligent algorithms cannot deal effectively massive dimensional station data. In this paper, distributed...
With the advance of software receiver, multipath estimation becomes a key issue for high accuracy positioning systems. It is crucial eliminating error and improving to estimate parameters. The accessible algorithms are usually designed Gaussian noise, their performances degrade dramatically in non-Gaussian since mean square criterion adopted. To tackle problem, new filter based on centered entropy (CEEC) proposed estimation. In filter, CEEC considered as performance index, which not limited...
The inherent randomness, fluctuation, and intermittence of photovoltaic power generation make it difficult to track the scheduling plan. To improve ability plan a greater extent, real-time charge discharge control method based on deep reinforcement learning is proposed. Firstly, energy storage hybrid system mathematical model are briefly introduced, tracking problem defined. Then, plans different days clustered into four scenarios by K-means clustering algorithm. mean, standard deviation,...
Quantitative myocardial tissue characterization with T1 and T2 parametric mapping can provide an accurate complete assessment of abnormalities across a broad range cardiomyopathies. However, current clinical tools rely predominantly on two-dimensional (2D) breath-hold sequences. Clinical adoption three-dimensional (3D) techniques is limited by long scan duration. The aim this study to develop validate time-efficient 3D free-breathing simultaneous sequence using multi-parametric...
Multipath is one of the dominant error sources for high-precision positioning systems, such as global navigation satellite systems. The minimum mean square criterion usually employed multipath estimation under assumption Gaussian noise. For non-Gaussian noise appeared in most practical applications, alternative solutions are required estimation. In this paper, a algorithm proposed based on entropy (MEE) or noises. A key advantage using MEE that it can minimize randomness signals; however,...
This paper presents a novel stochastic predictive tracking control strategy for nonlinear and non-Gaussian systems based on the single neuron controller structure in framework of information theory. Firstly, order to characterize randomness system, survival potential (SIP), instead entropy, is adopted formulate performance index, which not shift-invariant, i.e., its value varies with change distribution location. Then, optimal weights can be obtained by minimizing presented SIP criterion....