- Maritime Navigation and Safety
- Remote Sensing and LiDAR Applications
- Millimeter-Wave Propagation and Modeling
- Data Visualization and Analytics
- Remote-Sensing Image Classification
- Computer Graphics and Visualization Techniques
- Geographic Information Systems Studies
- Remote Sensing and Land Use
- Power Line Communications and Noise
- Data Management and Algorithms
- Underwater Acoustics Research
- Simulation and Modeling Applications
- Distributed and Parallel Computing Systems
- Advanced Computational Techniques and Applications
- Geological Modeling and Analysis
- Coastal and Marine Management
- Satellite Image Processing and Photogrammetry
- Advanced MIMO Systems Optimization
- Remote Sensing in Agriculture
- Atmospheric aerosols and clouds
- Advanced Image Fusion Techniques
- Coral and Marine Ecosystems Studies
- Advanced Vision and Imaging
- Traffic Prediction and Management Techniques
- Human Mobility and Location-Based Analysis
Beijing Jiaotong University
2010-2025
Chinese Academy of Sciences
2019-2025
Changchun Institute of Optics, Fine Mechanics and Physics
2019-2025
Shandong University of Science and Technology
2009-2024
Ministry of Natural Resources
2019-2022
Guangzhou Marine Geological Survey
2021
National Administration of Surveying, Mapping and Geoinformation of China
2020
China United Network Communications Group (China)
2004-2005
The local connection characteristics of convolutional neural network (CNN) are linked with the spatial correlation image pixels for water depth retrieval in this article. method has greater advantages and higher precision than traditional methods. Traditional remote sensing empirical models require manual extraction factors process is complex. This article proposes a model based on CNN, which uses different images four spectral bands, red, green, blue, near-infrared, to retrieve depth. In...
With the development of maritime economy, sea traffic is becoming more and crowded, accidents are also increasing. Research on search rescue decision-making technology cannot be delayed. This paper studies decision algorithm, based optimal theory. It analyzes three important concepts: Probability containment (POC), probability detection (POD), success (POS) involved in process. In this paper, calculation methods POC POD variables have been improved, rate has improved to some extent. Finally,...
Maritime search and rescue (SAR) decisions are the most important part of maritime SAR operations. In process making decisions, a key factor affecting efficiency success rate is how to quickly respond accidents develop an emergency response plan. At present, plans still mostly obtained through combination drift prediction models experience. There lack resource scheduling task assignment. The primary purpose this paper explore possibility using intelligent decision-making algorithm formulate...
The coastal zone plays an essential part in maintaining the balance of ecosystem and promoting development human society economy. It is significant to assess extent which Jiaozhou Bay withstands floods erosion during storms. exposure index (CEI) 1984, 2000 2019 was obtained by vulnerability model based on data including coastline, bathymetry terrain elevation. spatial distribution aggregation characteristics CEI were analyzed through autocorrelation analysis. results show that north coast...
In this letter, the underlying factor that leads to change of path loss exponent <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">n</i> is investigated, using a reasonable and reliable indirect demonstration based on simple two-ray model developed multiray model. It found first Fresnel zone affects . will only if situation obstacles in changes. The conclusion helps understand essences propagation.
With the development of deep learning in satellite remote sensing image segmentation, convolutional neural networks have achieved better results than traditional methods. In some full networks, number network layers usually increases to obtain features, but gradient disappearance problem occurs when deepens. Many scholars obtained multiscale features by using different calculations. We want structure while obtaining contextual information other means. This article employs self-attention...
Bike-sharing data have been a valuable source for urban transport research. While most studies focus on Origin-Destination (OD) of users' bike trips in bike-sharing researches, little has investigated mobility patterns from the perspective bikes. Bike can not only reflect human but also provide practical insights understanding system and assisting management. In particular, continuous movement information abundant spatiotemporal connection spatial context analysis without user ID...
Six-degree-of-freedom industrial robots, known for their low cost and high flexibility, have been extensively applied in optical processing. Precise pose control robot-based processing systems depends on the accurate calibration of tool coordinate system. However, robot magnetorheological finishing (Robot-MRF) systems, spherical shape polishing wheel poses significant challenges precisely identifying working point tool’s surface. Traditional methods, such as four-point or six-point...
How can robots learn dexterous grasping skills efficiently and apply them adaptively based on user instructions? This work tackles two key challenges: efficient skill acquisition from limited human demonstrations context-driven selection. We introduce AdaDexGrasp, a framework that learns library of single demonstration per selects the most suitable one using vision-language model (VLM). To improve sample efficiency, we propose trajectory following reward guides reinforcement learning (RL)...
Very high resolution (VHR) remote sensing images contain various multi-scale objects, such as large-scale buildings and small-scale cars. However, these objects cannot be considered simultaneously in the widely used backbones with large downsampling factor (e.g. VGG-like ResNet-like), resulting appearance of context aggregation approaches fusing low-level features attention-based modules. To alleviate this problem caused by factor, we propose a feature-selection high-resolution network...
As a very important parameter in link budget and channel modeling, the Ricean K factor viaduct cutting scenarios along high speed railway is estimated by using moment-based estimator. The practical measurement taken train at of more than 250 km/h. measured dist
Typhoons and other marine meteorological disasters often bring significant losses to human beings, their data are characterized by multiple sources scales, making traditional visualization methods unable accurately express the characteristics movement trends of disasters. To address above problems, this study proposes a typhoon dynamic method based on integration vector scalar fields. The uses ray casting visualize volume rendering data, hybrid interpolation improve efficiency, introduces...
Coral reef ecosystem is gradually being threatened, thus monitoring coral reefs using remote sensing of great significance. There are difficulties in identifying and classifying reefs. We propose a radiative transfer model with deep learning (RTDL) to improve the accuracy classification. This combines theory methods consider nonlinear fitting ability constraints physical model, which improves classification accuracy. utilizes ICESat-2 data obtain underwater topography photons (UTP) by active...