A5091806934- Traffic Prediction and Management Techniques
- Traffic control and management
- Transportation Planning and Optimization
- Time Series Analysis and Forecasting
- Machine Learning and ELM
- Energy Load and Power Forecasting
- Advanced MRI Techniques and Applications
- AI in cancer detection
- Advanced Neural Network Applications
- Systemic Lupus Erythematosus Research
- Radiomics and Machine Learning in Medical Imaging
- Anomaly Detection Techniques and Applications
- Advanced Image Fusion Techniques
- Image Enhancement Techniques
- Advanced Image Processing Techniques
- Medical Imaging Techniques and Applications
- Advanced Malware Detection Techniques
- Neural Networks and Applications
- Face and Expression Recognition
- Advanced Vision and Imaging
- Air Quality Monitoring and Forecasting
- Blockchain Technology Applications and Security
- Image and Signal Denoising Methods
- Medical Imaging and Analysis
- 3D Shape Modeling and Analysis
Hainan University
2023-2025
University of Electronic Science and Technology of China
2024-2025
Hong Kong Polytechnic University
2019-2024
Fudan University Shanghai Cancer Center
2024
Shantou University
2018-2023
Guizhou University
2023
Nanjing University of Aeronautics and Astronautics
2019-2023
Wenzhou Medical University
2006-2022
Key Laboratory of Guangdong Province
2022
Ministry of Education of the People's Republic of China
2022
Accurate and reliable traffic flow forecasting is of importance for urban planning mitigation congestion, it also the basis deployment intelligent management systems. However, constructing a reasonable robust model challenging task due to uncertainties nonlinear characteristics flow. Aiming at relationship affecting effect, PSO-ELM based on particle swarm optimization proposed short-term forecasting, which takes advantages search global optimal solution extreme learning machine fast deal...
Short‐term traffic flow forecasting is a fundamental and challenging task since it required for the successful deployment of intelligent transportation systems dramatically changing through time. This study presents novel hybrid dual Kalman filter (H‐KF 2 ) accurate timely short‐term forecasting. To achieve this, H‐KF first models propagation discrepancy between predictions traditional random walk model. By estimating posteriori state prediction errors both models, calibrated exploited to...
This paper presents a spatial-temporal deep learning network, termed ST-TrafficNet, for traffic flow forecasting. Recent methods highly relate accurate predetermined graph structure the complex spatial dependencies of flow, and ineffectively harvest high dimensional temporal features flow. In this paper, novel multi-diffusion convolution block constructed by an attentive diffusion bidirectional is proposed, which capable to extract precise potential dependencies. Moreover, stacked Long...
Abstract This study aimed to analyse cerebral grey matter changes in mild cognitive impairment (MCI) using voxel-based morphometry and diagnose early Alzheimer's disease deep learning methods based on convolutional neural networks (CNNs) evaluating these changes. Participants (111 MCI, 73 normal cognition) underwent 3-T structural magnetic resonance imaging. The obtained images were assessed morphometry, including extraction of matter, analyses statistical differences, correlation between...
Nowadays, accurate and efficient short-term traffic flow forecasting plays a critical role in intelligent transportation systems (ITS). However, due to the fact that is susceptible factors such as weather road conditions, data tend exhibit dynamic uncertainty nonlinearity, making construction of robust reliable model still challenging task. Aiming at this nonlinear complex problem, paper constructs hybrid optimization model, SSA-ELM, based on extreme learning machine by embedding sparrow...
Short-term traffic flow forecasting is an essential part of intelligent transportation systems. However, it challenging to model accurately due its rapid changes over time. The Kolmogorov–Arnold Network (KAN) has shown parameter efficiency with lower memory and computational overhead via spline-parametrized functions handle high-dimensional temporal data. In this paper, we propose unlock the potential network for by optimizing parameters a heuristic algorithm. gravitational search algorithm...
Short-term traffic flow forecasting is a fundamental and challenging task due to the stochastic dynamics of flow, which often imbalanced noisy. This paper presents sample-rebalanced outlier-rejected k-nearest neighbor regression model for short-term forecasting. In this model, we adopt new metric evolutionary patterns, reconstruct balanced training sets by relative transformation tackle imbalance issue. Then, design hybrid that considers both local global information address limited size...
Accurate and timely short-term traffic flow forecasting plays a key role in intelligent transportation systems, especially for prospective control. For the past decade, series of methods have been developed forecasting. However, due to intrinsic stochastic evolutionary trend, accurate remains challenging. In this paper, we propose noise-immune long memory (NiLSTM) network forecasting, which embeds loss function deduced by maximum correntropy into (LSTM) network. Different from conventional...
Accurate and timely short‐term traffic flow forecasting is a critical component for intelligent transportation systems. However, it quite challenging to develop an efficient robust model due complex non‐linear data pattern of flow. Support vector regression (SVR) has been widely employed in time series prediction problems. the lack knowledge choice hyper‐parameters SVR leads poor accuracy. In this study, authors propose hybrid combining gravitational search algorithm (GSA) model. The GSA...
Multimodality sentiment analysis in the social Internet of Things is a developing field, which basic to empathetic mechanisms, affective computing, and artificial intelligence. Current works this domain do not explicitly consider influence contextual information fusion based on correlation coefficient memory network with branch structure for analysis. Unlike present works, article presents hierarchical self-attention (H-SATF) model capturing better among utterances, temporal convolutional...
Color correction and enhancement for underwater images is challenging due to attenuation scattering. The often have low visibility suffer from color bias. This paper presents a novel method based on filter array (CFA) an Retinex with dense pixels adaptive linear histogram transformation degraded color-biased images. For any digital image in the RGB space, which captured by camera CFA, their values are dependent coupled because of interpolation process. So we try compensate red channel green...
Stroke is the second leading cause of death globally and most common severe disability. Several barriers need to be addressed more effectively treat stroke, including efficient delivery therapeutic agents, rapid release at infarct site, precise imaging drug distribution monitoring. The present study aimed develop a bio-responsive theranostic nanoplatform with signal-amplifying capability deliver rapamycin (RAPA) ischemic brain tissues visually monitor distribution. A pH-sensitive RAPA-loaded...
Extensive traffic flow data are received from the loop detector networks every second, which requires us to develop an effective and efficient algorithm predict future flow. However, dynamic conditions on a road not just influenced by sequential patterns in temporal dimension, but also other roadways spatial dimension. Although many successful models have been developed previous studies forecast flows, most of them shortcomings modeling dependencies. In this article, we focus...