A5101921924- Land Use and Ecosystem Services
- Remote-Sensing Image Classification
- Flood Risk Assessment and Management
- Remote Sensing and Land Use
- Remote Sensing in Agriculture
- Muscle activation and electromyography studies
- Hydrology and Watershed Management Studies
- Environmental Changes in China
- Geophysics and Gravity Measurements
- Hydrological Forecasting Using AI
- Balance, Gait, and Falls Prevention
- Environmental and Agricultural Sciences
- Climate change and permafrost
- Soil Moisture and Remote Sensing
- Gamma-ray bursts and supernovae
- Remote Sensing and LiDAR Applications
- Environmental Monitoring and Data Management
- Time Series Analysis and Forecasting
- Gait Recognition and Analysis
- Advanced Image Fusion Techniques
- Water Resources and Management
- Glaucoma and retinal disorders
- Aquatic Ecosystems and Biodiversity
- Automated Road and Building Extraction
- Lipid metabolism and disorders
Yunnan University
2023-2024
Zhejiang A & F University
2020-2022
Shenzhen University
2015-2021
Ministry of Natural Resources
2020-2021
Tongji University
2014-2020
Xiamen University of Technology
2016
Xiamen University
2014-2015
Chalmers University of Technology
2000-2002
Water bodies are a fundamental element of urban ecosystems, and water mapping is critical for landscape planning management. Remote sensing has increasingly been used in rural areas; however, when applied to areas, this spatially- explicit approach challenging task due the fact that often small size spectral confusion common between complex features environment. indexes most method extraction at pixel level. More recently, mixture analysis (SMA) widely employed analyzing environment subpixel...
The successful launch of the Landsat 8 satellite continues Earth observation series, which has been taking place for nearly 40 years. With increase in band number and improved spectral range compared with previous imagery, it will be possible to expand application new imagery. purpose this study is explore water extraction based on Operational Land Imager (OLI) According specific inland conditions (clear water, turbid eutrophic water), a highly adaptable indices are assessed using OLI...
Earth's surface water plays an important role in the global cycle, environmental processes, and human society, it is necessary to dynamically capture distribution extent of on Earth. However, due high complexity environment Earth, current mapping methods are limited applicability precision. In this study, explore automatic applicable model for mapping, particularly regions with highly heterogenous backgrounds, we adopted state-of-the-art deep learning techniques structured a new model,...
Future land use and cover change (LUCC) simulations play an important role in providing fundamental data to reveal the carbon cycle response of forest ecosystems LUCC. Subtropical forests have great potential for sequestration, yet their future dynamics under natural human influences are unclear. Zhejiang Province China is distribution area subtropical forests. For management, it significance explore dynamic changes Zhejiang. As a popular LUCC spatial simulation model, cellular automata (CA)...
The accuracy of training samples used for data classification methods, such as support vector machines (SVMs), has had a considerable positive impact on the results urban area extractions. To improve built-up extractions, this paper presents sample-optimized approach classifying using combination Defense Meteorological Satellite Program-Operational Linescan System (DMSP-OLS) nighttime light data, Landsat images, and GlobeLand30, which is 30-m global land cover product. proposed consists...
Extracting surface land-cover types and analyzing changes are among the most common applications of remote sensing. One basic tasks is to identify map water boundaries. Spectral indexes have been successfully used in extraction bodies multispectral images. However, directly applying a index method hyperspectral images disregards abundant spectral information involves difficulty selecting appropriate bands. It also challenge for distinguish from shadowed regions. The purpose this study...
In this paper, we calibrate the luminosity relation of gamma-ray bursts (GRBs) from an Artificial Neural Network (ANN) framework for reconstructing Hubble parameter \unboldmath{$H(z)$} latest observational data (OHD) obtained with cosmic chronometers method in a cosmology-independent way. We consider physical relationships between to introduce covariance matrix and KL divergence construct loss function Amati ($E_{\rm p}$--$E_{\rm iso}$) by selecting optimal ANN model A219 sample J220 at low...
Accurate land use/land cover (LULC) mapping over a large area is essential to environmentally sustainable development. Recently, the Chinese government established new national economic zone called Xiongan New Area, and along with upcoming large-scale urban construction, this will inevitably experience dramatic LULC change, which threaten local ecological balance. In article, we proposed two-stage approach for in Area ahead of forthcoming dense construction. The first stage obtain base-class...
The surface water in the Qinghai–Tibet Plateau (QTP) region has undergone dramatic changes recent decades. To capture dynamic information, many satellite imagery-based methods have been proposed. However, these are still limited terms of automation and accuracy thus prevent studies large-scale QTP regions. In this study, we developed a new fully automatic method for accurate mapping by using Sentinel-1 synthetic aperture radar (SAR) imagery convolutional networks (ConvNets). Specifically,...
The application of deep learning techniques, especially convolutional neural networks (DCNNs), in the intelligent mapping very high spatial resolution (VHSR) remote sensing images has drawn much attention community. However, fragmented distribution urban land use types and complex structure forests bring about a variety challenges for extraction forests. Based on DCNN algorithm, this study proposes novel object-based U-net-DenseNet-coupled network (OUDN) method to realize accurate proposed...
The pathological condition in a degenerative knee joint may be assessed by analyzing the vibroarthrographic signals. With severity level of disorders evaluated computational methods, unnecessary imaging examination or open surgery can prevented. In present study, we used k-nearest neighbor (k-NN) algorithm, type lazy learning approach, to classify signals collected from healthy subjects and symptomatic patients with disorders. representative features form factor variance mean-square values,...
Water bodies are a fundamental element of urban ecosystems, and water mapping is critical for landscape planning management. Remote sensing has increasingly been used in rural areas; especially, hyperspectral remote image characterized with rich spectrum information provide greater potential high-accuracy land cover classification, however, the hundreds bands contained also poses huge burden on data processing. In this study, aims extraction densely built area, we proposed fast method based...
Moderate spatial resolution (MSR) satellite images, which hold a trade-off among radiometric, spectral, and temporal characteristics, are extremely popular data for acquiring land cover information. However, the low accuracy of existing classification methods MSR images is still fundamental issue restricting their capability in urban mapping. In this study, we proposed hybrid convolutional neural network (H-ConvNet) improving mapping with Sentinel-2 images. The H-ConvNet was structured two...
A hierarchical method for subpixel surface water mapping accounting the high spectral heterogeneity of urban materials is proposed in this paper. Specifically, we first applied index (WI) remote sensing image classification at pixel level, afterwards, land, water, and land-water mixture can be extracted automatically. Then analysis (SMA) to mixed pixels fraction estimation level. To obtaining most representative endmembers SMA, designed an adaptive iterative endmember selection based on...
In this paper, we derive forward autoregressive models to describe the stochastic process underlying stride interval series related idiopathic Parkinson's disease. The parameters of model that specify pole locations in complex z-plane were used as dominant features for separation gait healthy subjects and patients with Based on parameters, linear discriminant analysis support vector machines can provide classification accurate rates over 74% area larger than 0.8 under receiver operating...