A5081235198- Geographic Information Systems Studies
- Automated Road and Building Extraction
- Data Management and Algorithms
- Human Mobility and Location-Based Analysis
- Traffic Prediction and Management Techniques
- Remote Sensing and LiDAR Applications
- Remote-Sensing Image Classification
- Transportation Planning and Optimization
- Advanced Computational Techniques and Applications
- Remote Sensing and Land Use
- Complex Network Analysis Techniques
- Image Retrieval and Classification Techniques
- 3D Modeling in Geospatial Applications
- Advanced Image and Video Retrieval Techniques
- Fiber-reinforced polymer composites
- Quantum Dots Synthesis And Properties
- Groundwater and Watershed Analysis
- Advanced Sensor and Energy Harvesting Materials
- Image Processing and 3D Reconstruction
- Land Use and Ecosystem Services
- Advanced materials and composites
- Geoscience and Mining Technology
- Network Security and Intrusion Detection
- Hydrology and Watershed Management Studies
- Geomechanics and Mining Engineering
Wuhan University
2016-2025
Shanghai University
2009-2024
Tongji University
2024
University of Hong Kong
2014-2024
Research Institute of Petroleum Exploration and Development
2024
China National Petroleum Corporation (China)
2024
Hong Kong University of Science and Technology
2024
Wuyi University
2019-2023
Shanghai Architectural Design & Research Institute
2023
Shanghai Xiandai Architectural Design
2023
Machine learning methods, specifically, convolutional neural networks (CNNs), have emerged as an integral part of scientific research in many disciplines. However, these powerful methods often fail to perform pattern analysis and knowledge mining with spatial vector data because most cases, such are not underlying grid-like or array structures but can only be modeled graph structures. The present study introduces a novel convolution by converting it from the vertex domain into point-wise...
The shape of a geospatial object is an important characteristic and significant factor in spatial cognition. Existing representation methods for vector-structured objects the map space are mainly based on geometric statistical measures. Considering that complicated cognitively related, this study develops learning strategy to combine multiple features extracted from its boundary obtain reasonable representation. Taking building data as example, first models using graph structure extracts...
The automatic classification of urban functional regions is vital for planning and governance. current methods mainly rely on single remote sensing image data or social data. However, these imagery-based have the disadvantage capturing high-level socioeconomic features, whereas information from alone rarely contains morphological features. To overcome limitations, it necessary to combine multisource functionalities. This study presents an ensemble method that combines vector-based buildings...
As a basic and significant operator in map generalization, polyline simplification needs to work across scales. Perkal's ε-circle rolling approach, which circle with diameter ε is rolled on both sides of the so that small bend features can be detected removed, considered as one few scale-driven solutions. However, envelope computation, key part this method, has been difficult implement. Here, we present computational method implements proposal. To simulate effects circle, Delaunay...
Abstract Flexible sensors are required to be lightweight, compatible with the skin, sufficiently sensitive, and easily integrated extract various kinds of body vital signs during continuous healthcare monitoring in daily life. For this, a simple low-cost flexible temperature force sensor that uses only two carbon fiber beams as sensing layer is reported this work. This simple, can not monitor skin changes real time but also most pulse waves, including venous from parts human body. A...
Drainage pattern recognition (DPR) is a classic and challenging problem in hydrographic system analysis, topographical knowledge mining, map generalization. An outstanding issue for traditional DPR methods that the rules used to extract patterns based on certain geometric measures are limited, not accessing effects of manual recognition. In this study, graph convolutional network (GCN) was introduced DPR. First, dual drainage built channel connection hierarchical structure after constructing...
Detecting interchanges in road networks benefit many applications, such as vehicle navigation and map generalization. Traditional approaches use manually defined rules based on geometric, topological, or both properties, thus can present challenges for structurally complex interchange. To overcome this drawback, we propose a graph-based deep learning approach interchange detection. First, model the network graph which nodes represent segments, edges their connections. The proposed computes...
Owing to their limited accuracy and narrow applicability, current antimicrobial peptide (AMP) prediction models face obstacles in industrial application. To address these limitations, we developed improved an AMP model using Comparing Optimizing Multiple DEep Learning (COMDEL) algorithms, coupled with high-throughput screening method, finally reaching of 94.8% test 88% experiment verification, surpassing other state-of-the-art models. In conjunction COMDEL, employed the phage-assisted...
For point clusters, the conflict and crowding of map symbols is an inevitable problem during transition from large to small scales. The cartographic generalization involved in this as a spatial decision-making process usually related analysis context, choice abstraction operators, judgment resulting data quality. rules summarized by traditional methods require manual setting conditions or thresholds sometimes encounter special cases that make it difficult directly match certain integrate...
OpenStreetMap (OSM) has been playing increasingly important roles in location-based services, urban planning, tourism, and disaster management. In this paper, we examine a fundamental issue: can OSM reliably provide its services the long run? To address sustainable issue, propose (1) inner cycle of career stages (i.e. activity levels) for monitoring status (2) critical mass theory identifying driving factors that make sustain. These new tools are used to track individual trajectories editing...
In map generalization, the displacement operation attempts to resolve proximity conflicts guarantee legibility. Owing limited representation space, may occur between both same and different features under contexts. A successful should settle multiple conflicts, suppress generation of secondary after moving some objects, preserve distribution patterns. The effect can be understood as a force that pushes related objects away with properties propagation distance decay. This study borrows idea...
Research has developed numerous algorithms to simplify building data. Each algorithm strengths and weaknesses in addressing shape characteristics, but no single can appropriately all buildings. This study proposes a hybrid approach that identifies the best simplified representation of among four existing algorithms. The proposed applies generate simplification candidates. With backpropagation neural network, an evaluator is built through supervised learning based on measurements describing...
Traffic prediction is a fundamental and vital task in Intelligence Transportation System (ITS), but it very challenging to get high accuracy while containing low computational complexity due the spatiotemporal characteristics of traffic flow, especially under metropolitan circumstances. In this work, new topological framework, called Linkage Network, proposed model road networks present propagation patterns flow. Based on Network model, novel online predictor, named Graph Recurrent Neural...
Identifying the spatial configurations of buildings and grouping them reasonably is an important task in cartography. This study developed a approach using graph deep learning by integrating multiple cognitive features manual cartographic experiences. Taking building center points as nodes, adjacent were organized which variables including size, orientation, shape defined for each node. Then, model combining convolution neural network was designed to analyse modelled graph. The groups used...
Similarity measurement has been a prevailing research topic in geographic information science. Geometric similarity scaling transformation (GSM_ST) is critical to ensure spatial data quality while balancing detailed with distinctive features. However, GSM_ST an uncertain problem due subjective cognition, global and local concerns, geometric complexity. Traditional rule-based methods considering multiple consistent conditions require adjustments characteristics weights, leading poor...
Polyline and building simplification remain challenging in cartography. Most proposed algorithms are geometric-based rely on specific rules. In this study, we propose a deep learning approach to simplify polylines buildings based graph autoencoder (GAE). The model receives the coordinates of line vertices as inputs obtains simplified representation by reconstructing original with fewer through pooling, which convolution Fourier transform is used for layer-by-layer feature computation. By...
This paper presents a novel data fusion framework to address the challenges of extracting valuable information from big characterized by its "5V" attributes. The integrates Dempster-Shafer Evidence Theory (DSET) for handling fuzzy and Information optimizing process, resulting in comprehensive robust approach. A new algorithm preprocessing classification is proposed, utilizing K-means clustering differentiate between precise data. Fuzzy processed using DSET, while Bhattacharyya Jensen-Shannon...