A5103119407- Traffic Prediction and Management Techniques
- Human Mobility and Location-Based Analysis
- Urban Transport and Accessibility
- Infrastructure Maintenance and Monitoring
- Anomaly Detection Techniques and Applications
- Energy Load and Power Forecasting
- Elevator Systems and Control
- Social Robot Interaction and HRI
- Adventure Sports and Sensation Seeking
- Vehicle Routing Optimization Methods
- Advanced Sensor and Control Systems
- Transportation Planning and Optimization
- Data-Driven Disease Surveillance
- Robotics and Automated Systems
- Music and Audio Processing
- Emotion and Mood Recognition
- Smart Cities and Technologies
- Neural Networks and Applications
- Advanced Image Processing Techniques
- Transportation and Mobility Innovations
- Data Management and Algorithms
- Facility Location and Emergency Management
- Speech and Audio Processing
- Advanced Algorithms and Applications
- Image and Signal Denoising Methods
Jilin University
2017-2023
Beijing University of Technology
2023
Jilin Medical University
2023
Wuhan University of Technology
2022
Urban traffic demand distribution is dynamic in both space and time. A thorough analysis of individuals’ travel patterns can effectively reflect the dynamics a city. This study aims to develop an analytical framework explore spatiotemporal characteristics community structure shaped by travel, which analyzed empirically New York City. It uses spatial statistics graph-based approaches quantify behaviors generate previously unobtainable insights. Specifically, people primarily for commuting on...
Based on improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) algorithm and kernel online sequential extreme learning machine (KOSELM) algorithm, a new hybrid short-term traffic flow prediction model (ICEEMDAN-KOSELM-ARIMA) for signalized intersections is proposed according to the current historical data. First, time series are decomposed by ICEEMDAN purpose of improving accuracy. Several intrinsic functions could be obtained process. Then, permutation...
This article presents a Location‐Routing Problem (LRP) model to assist decision makers in emergency logistics. The attempts consider the relationship between location of warehouses and delivery routes order maximize rescue efficiency. objective function minimization time cost is established single‐stage LRP considering different scenarios. hybrid self‐adaptive bat algorithm (HSABA) an improved nature‐inspired for solving this model, hard optimization problem. HSABA with self‐adaptation...
As an emerging type of spatio-temporal big data based on positioning technology and navigation devices, vehicle-based crowdsourcing has become a valuable trajectory resource. However, been collected by non-professionals with multiple measurement terminals, resulting in certain errors collection. In these cases, to minimize the impact outliers obtain relatively accurate data, it is crucial detect clean outliers. This paper proposes efficient outlier detection (CTOD) method that detects from...
In the real world, traffic scenes are complex and contain intricate road networks, which makes shortest path computation on large-scale networks a challenging task. Existing research has concentrated small-scale urban or grid maps. practical scenarios, however, we often faced with seeking paths networks. For this reason, it is imperative to develop efficient searching algorithms, as offers significant savings in time resources. To tackle issue, paper proposes two improved A* namely Weighted...
In this paper, we investigate an image restoration method based on a deep reinforcement learning framework, aiming to recover low quality images high images. The general learning-based approach trains single large network accomplish specific task, which is difficult handle when facing mixed distorted To address problem, propose the residual Double DQN algorithm, introduces idea of residuals into framework. agent learns policy select appropriate actions from action set gradually restore...
Speech emotion recognition (SER) plays a crucial role in classifying emotional information conveyed through audio signals, providing more accurate and convenient solutions for human-computer interaction, analysis other fields. The recent research has focused on training Transformer-based models human-annotated datasets to capture long-range dependencies by modeling fixed-scale feature representations, processing time-varying spectral features as images. However, extracting efficient robust...
A large amount of real-time traffic data supports the processing requirements state discrimination and prediction.Therefore, accurate information can be grasped for effective detection outliers.In this paper, an optimized AdaBoost model screening abnormal samples is proposed based on multi-source features detected data.Considering unbalanced characteristics data, by cost-sensitive method, which avoids problem that classification performance degraded non-equilibrium data.The accuracy, false...
With the development of construction industry, work more and tends to mechanization, automation, intelligence. The appearance handling robots greatly improves efficiency reduces cost. Even though great efforts have been made on motion pattern recognition for robots, existing methods still some shortcomings. Previous studies are all used recognition, each input is processed separately, which cuts off time continuity between latter data former data. To this end, in paper, we first emphasize...