A5108064895- Machine Learning and ELM
- Remote-Sensing Image Classification
- Domain Adaptation and Few-Shot Learning
- Advanced Algorithms and Applications
- Face and Expression Recognition
- Atmospheric chemistry and aerosols
- Reinforcement Learning in Robotics
- Optical Systems and Laser Technology
- Remote Sensing and Land Use
- Advanced SAR Imaging Techniques
- Gene expression and cancer classification
- Advanced Image Fusion Techniques
- Air Quality and Health Impacts
- Advanced Measurement and Detection Methods
- Electric Motor Design and Analysis
- Adaptive Dynamic Programming Control
- Advanced Sensor and Control Systems
- Fault Detection and Control Systems
- Advanced Computational Techniques and Applications
- Cancer-related molecular mechanisms research
- Bioinformatics and Genomic Networks
- Radar Systems and Signal Processing
- Industrial Technology and Control Systems
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Neural Networks and Applications
China University of Mining and Technology
2016-2025
Ministry of Education of the People's Republic of China
2021-2025
Beijing Normal University
2010-2025
University of California, Riverside
2020-2025
State Key Joint Laboratory of Environment Simulation and Pollution Control
2011-2024
Peking University
2011-2024
Suzhou University of Science and Technology
2022-2024
National University of Defense Technology
2015-2024
Qiqihar University
2024
State Grid Corporation of China (China)
2022-2024
Blur is a key determinant in the perception of image quality. Generally, blur causes spread edges, which leads to shape changes images. Discrete orthogonal moments have been widely studied as effective descriptors. Intuitively, can be represented using discrete since noticeable affects magnitudes an image. With this consideration, paper presents blind evaluation algorithm based on Tchebichef moments. The gradient blurred first computed account for shape, more representation. Then divided...
The Industrial Internet of Things (IIoT) has developed rapidly in recent years. Private blockchains with decentralization, flexible rules, and good privacy protection can be applied the IIoT to process massive data tackle security problem. However, scalability blockchain places a restriction on IIoT. Accordingly, this article proposes an improved algorithm based Two_Arch2 improve decentralization while reducing latency cost blockchain. By integrating private theory simultaneously considering...
Abstract Motivation Accumulating clinical observations have indicated that microbes living in the human body are closely associated with a wide range of noninfectious diseases, which provides promising insights into complex disease mechanism understanding. Predicting microbe–disease associations could not only boost diagnostic and prognostic, but also improve new drug development. However, little efforts been attempted to understand predict on large scale until now. Results In this work, we...
For CareBeijing‐2006, two sites were established in urban and suburban regions of Beijing summer 2006. Observations O 3 its precursors together with meteorological parameters at both are presented. Gross ozone production rate P (O ) sensitivity to nitric oxides (NO x volatile organic compounds (VOCs) investigated using an observation‐based photochemical box model (OBM). varied from nearly zero 120 50 ppb h −1 for sites, respectively. These rates greater than the accumulation observed oxidant...
To improve the accuracy of multi-step wind power forecast, a variational mode decomposition-long short-term memory (VMD-LSTM) forecast method is proposed. Firstly, decomposition adopted to decompose data into three constituent modes, named as long-term component, fluctuation component and random component. Secondly, long network utilised deeply learn characteristics modes. Profit from its unique forget gate structure, association with time series learned build model. Finally, ELIA NERL are...
Assessment and characterization of human colon microbiota is now a major research area in diseases, including patients with hepatitis B liver cirrhosis (HBLC).We recruited 120 HBLC healthy controls. The fecal microbial community functions the two groups were analyzed using high-throughput Solexa sequencing complete metagenomic DNA bioinformatics methods.Community metabolism-wide changes 20 controls observed compared. A negative correlation was between Child-Turcotte-Pugh scores Bacteroidetes...
This paper presents the development of a uniform polarimetric matrix rotation theory in domain along radar line sight for synthetic aperture (PolSAR) data interpretation. The representation each coherency element is sinusoidal function domain. A set oscillation parameters, including amplitude, center, angular frequency, and initial angle, proposed to fully characterize scattering behavior angle stationary null minimization/maximization angles, derived indicate specific states property....
Broad learning system (BLS) is a novel neural network with effective and efficient ability. BLS has attracted increasing attention from many scholars owing to its excellent performance. This article proposes weighted (WBLS) based on tackle the noise outliers in an industrial process. WBLS provides unified framework for easily using different methods of calculating penalty factor. Using factor constrain contribution each sample modeling, normal abnormal samples were allocated higher lower...
Accumulating experimental studies have indicated the influence of lncRNAs on various critical biological processes as well disease development and progression. Calculating lncRNA functional similarity is high value in inferring functions identifying potential lncRNA-disease associations. However, little effort has been attempt to measure among a large scale. In this study, we developed Fuzzy Measure-based LNCRNA SIMilarity calculation model (FMLNCSIM) based assumption that functionally...
The purpose of this study was to compare the prevalence self-reported mental health factors, service use, and unmet needs across 4 largest ethnic groups in Ontario, Canada: white, South Asian, Chinese, black groups.The population derived from Canadian Community Health Survey, using a cross-sectional sample 254,951 residents living Canada, between 2001 2014. Age- sex-standardized estimates for were calculated each overall by sociodemographic characteristics.We found that physician-diagnosed...
Deep reinforcement learning (RL) comprehensively uses the psychological mechanisms of "trial and error" "reward punishment" in RL as well powerful feature expression nonlinear mapping deep learning. Currently, it plays an essential role fields artificial intelligence machine Since agent needs to constantly interact with its surroundings, Q network (DQN) is inevitably faced need learn numerous parameters, which results low efficiency. In this paper, a multisource transfer double DQN (MTDDQN)...
Recently, deep learning-based methods have drawn increasing attention in hyperspectral imagery (HSI) classification, due to their strong nonlinear mapping capability. However, these suffer from a time-consuming training process because of many network parameters. In this paper, the concept broad learning is introduced into HSI classification. Firstly, make full use abundant spectral and spatial information imagery, hierarchical guidance filtering performing on original get its...
MicroRNAs (miRNAs), which play critical roles in gene regulatory networks, have emerged as promising diagnostic and prognostic biomarkers for human cancer. In particular, circulating miRNAs that are secreted into circulation exist remarkably stable forms, enormous potential to be leveraged non-invasive early cancer detection. Novel user-friendly tools desperately needed facilitate data mining of the vast amount miRNA expression from The Cancer Genome Atlas (TCGA) large-scale profiling...
As a class of randomized learner model, stochastic configuration networks (SCNs) have been successfully applied in few data analytics tasks. Given the industrial big modeling tasks, however, original SCNs potentially lead to excessive training time. To this end, article extends hybrid parallel version, termed (HPSCNs). In learning algorithm, two are synchronously constructed. The difference between them is that one employs point-incremental and another adopts block-incremental algorithm....
Abstract In the microgrid, virtual synchronous generator technology can significantly enhance anti‐interference characteristics of system frequency and bus voltage, as well solve problems insufficient damping low inertia. However, active power oscillation caused by fluctuations grid faults threaten stable operation seriously. Therefore, for an alternating current (AC) microgrid multi‐virtual (VSG) parallel system, improved control algorithm based on a fuzzy inference is proposed, which...
Abstract. The Pearl River Delta (PRD) region in South China is faced with severe ambient tropospheric ozone (O3) pollution autumn and summer, which mostly coincides the occurrence of typhoons above Northwest Pacific. With increasingly O3 PRD under influence typhoons, it necessary to gain a comprehensive understanding impact on transport, production accumulation for efficient reduction. In this study, we analysed general via systematic comparisons meteorological conditions, processes sources...
Atrial fibrillation (AF) is an increasing medical burden worldwide, and its pathological manifestations are atrial tissue remodeling low-pressure fibrosis. Due to the inherent defects of image data acquisition systems, high-resolution cardiac magnetic resonance imaging (CMRI) faces many problems. In response these problems, we propose Progressive Feedback Residual Attention Network (PFRN) for CMRI super-resolution. Specifically, directly perform feature extraction on low-resolution images,...