A5031035914- Precipitation Measurement and Analysis
- Meteorological Phenomena and Simulations
- Advanced Image and Video Retrieval Techniques
- Video Surveillance and Tracking Methods
- Image Retrieval and Classification Techniques
- Face and Expression Recognition
- Human Pose and Action Recognition
- Climate variability and models
- Soil Moisture and Remote Sensing
- Membrane Separation and Gas Transport
- Membrane Separation Technologies
- Market Dynamics and Volatility
- Blockchain Technology Applications and Security
- Robotic Path Planning Algorithms
- Stock Market Forecasting Methods
- Membrane-based Ion Separation Techniques
- Anomaly Detection Techniques and Applications
- Optimization and Search Problems
- Metaheuristic Optimization Algorithms Research
- Video Analysis and Summarization
- Gene expression and cancer classification
- Time Series Analysis and Forecasting
- Advanced Neural Network Applications
- Flood Risk Assessment and Management
- Underwater Acoustics Research
Sun Yat-sen University
2024-2025
National Sun Yat-sen University
2021-2024
Colorado State University
2023
Nanjing Normal University
2023
Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application
2023
Ocean University of China
2022
Beijing University of Technology
2012-2018
Maebashi Institute of Technology
2012
Developing region-specific radar quantitative precipitation estimation (QPE) products for South China (SC) is crucial due to its unique climate and complex terrain over there. Deep learning (DL) has emerged as a promising avenue QPE, especially graph neural networks (GNNs). Many studies have tested the DL models in but virtually no evaluated performance of different intensities, types, or organizations. Moreover, limited attention been given whether DL-based methods can mitigate QPE...
Feature selection is an effective machine learning method for reducing dimensionality, removing irrelevant features, increasing accuracy, and improving result comprehensibility. However, many existing feature methods are incapable high dimensional data because of their time complexity, especially wrapper algorithms. In this work, a fast sequential algorithm (AP-SFS) proposed based on affinity propagation clustering. AP-SFS divides the original space into several subspaces by cluster...
High-resolution radar rainfall data have great potential for predictions up to 6 h ahead (nowcasting); however, conventional extrapolation approaches based on in-built physical assumptions yield poor performance at longer lead times (3–6 h), which limits their operational utility. Moreover, atmospheric factors in estimate errors are often ignored. This study proposed a nowcasting method that attempts achieve accurate of using long short-term memory (LSTM) networks. Atmospheric conditions...
In this study, the fouling behavior of PES ultrafiltration (UF) membrane with different DOM fractions including bovine serum albumin (BSA), sodium alginate (SA) and humic acid (HA) was systematically investigated. The result showed that mechanism HA cake formation while BSA SA caused by both pore blocking due to particle size. Moreover, became more severe increase feed concentration TMP it could be accurately described cake-complete model. resistance for larger than BSA, whereas followed...
Cryptocurrency, a novel digital asset within the blockchain technology ecosystem, has recently garnered significant attention in investment world. Despite its growing popularity, inherent volatility and instability of cryptocurrency investments necessitate thorough risk evaluation. This study utilizes Autoregressive Moving Average (ARMA) model combined with Generalized Conditionally Heteroscedastic (GARCH) to analyze three major cryptocurrencies-Bitcoin (BTC), Ethereum (ETH), Binance Coin...
In this study, we propose a deep learning nowcasting method called multivariate channel transformer U-net (MCT If-net). It incorporates the into basic structure. Three dual-polarization radar variables, i.e., reflectivity ( <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$Z$</tex> ), specific differential phase xmlns:xlink="http://www.w3.org/1999/xlink">$K_{dp}$</tex> and xmlns:xlink="http://www.w3.org/1999/xlink">$Z_{dr}$</tex> are used as...
The growing popularity of online sports and exercise necessitates effective methods for evaluating the quality executions. Previous action assessment methods, which relied on labeled scores from motion videos, exhibited slightly lower accuracy discriminability. This limitation hindered their rapid application to newly added exercises. To address this problem, paper presents an unlabeled Multi-Dimensional Exercise Distance Adaptive Constrained Dynamic Time Warping (MED-ACDTW) method...
Distance measure is quite important for pattern recognition. Utilizing invariance in image data, tangent distance very powerful classifying handwritten digits. For this a set of invariant transformations must be known priori. But many practical problems, it difficult to know these transformations. In paper, an algorithm proposed approximate the exclusively from data. By virtue ideas arising manifold learning, needs no prior and can applied more classification problems. k-nearest neighbor...
Distance measure is quite important for pattern recognition. Utilizing invariance in image data, tangent distance very powerful classifying handwritten digits. For this a set of invariant transformations must be known priori. But many practical problems, it difficult to know these transformations. In paper, an algorithm proposed approximate the exclusively from data. By virtue ideas arising manifold learning, needs no prior and can applied more classification problems. k-nearest neighbor...
The task of nowcasting by deep learning using multivariate, rather than just reflectivity, is limited poor interpretability. previous experiment designed MCT (Multivariate Channel Transformer), a model capable with dual-polarization radar data. Four analytical methods are to further explore the contribution polarization parameters: (i) Case studies different meteorological processes. (ii) A permutation test ranking significance each variable. (iii) Visualization feature maps obtained forward...
Research on nowcasting through dual-polarization weather radar data using deep learning approach is rare but worth exploring. This paper lightens a previous work, the MCT (Multivariate Channel Transformer) model, which leads to design of MSF Swin Fusion) model. The commonalities between two are as follows: one hand, both fuses several observables including reflectivity (Z), specific differential phase (K <inf xmlns:mml="http://www.w3.org/1998/Math/MathML"...