A5101912038- Urban Heat Island Mitigation
- Land Use and Ecosystem Services
- Cryospheric studies and observations
- Urban Green Space and Health
- Ocean Waves and Remote Sensing
- Air Quality and Health Impacts
- Coastal and Marine Dynamics
- Oceanographic and Atmospheric Processes
- Air Quality Monitoring and Forecasting
- Remote Sensing and Land Use
- Advanced Image Fusion Techniques
- Climate variability and models
- Traditional Chinese Medicine Studies
- Remote-Sensing Image Classification
- Remote Sensing and LiDAR Applications
- Climate Change and Health Impacts
- Meteorological Phenomena and Simulations
- Flood Risk Assessment and Management
- Remote Sensing in Agriculture
- Impact of Light on Environment and Health
Chinese Academy of Sciences
2016-2025
Institute of Computing Technology
2025
University of Chinese Academy of Sciences
2007-2020
Guangzhou Institute of Geography
2016-2020
Guangzhou Institute of Geochemistry
2016-2020
Abstract Objective: This study addresses the limitations of existing traditional Chinese medicine (TCM) constitution identification techniques for elderly by proposing an intelligent method aimed at enhancing accuracy, standardization, and formalization process. Materials Methods: Leveraging data from images tongue, face, pulse, this introduced four image classification models: EfficientNetV2, MobileViT, Vision Transformer, Swin Transformer. A comparative experimental approach was employed...
Land surface characteristics, including soil type, terrain slope, and antecedent moisture, have significant impacts on runoff during heavy precipitation in highly urbanized areas. In this study, a Linear Spectral Mixture Analysis (LSMA) method is modified to extract high-precision impervious surface, vegetation, fractions. the LSMA method, representative endmembers are first selected by combining high-resolution image from Google Earth; unmixing results of then post-processed reduce errors...
Because of the low spatial resolution MODIS data, it is important to decompose mixed pixels retrieve component temperatures using thermal infrared bands. Grasslands with different coverage conditions are prominent in area under study. simple vegetation structure, radiation less influenced by shade. If internal structure parts pixel ignored, total emitted approximately sum each part pixel, weighted according percentage part. Vegetation/soil based on sub‐pixel scale inverted a constrained...
Accurate estimations of impervious surface areas are essential for urban planning development. Linear spectral mixture analysis (LSMA) is commonly adopted to extract the (IS) fraction in a mixed pixel at subpixel scale. However, owing errors spectra pure pixels selected from remote sensing images, incorrect fractions different land cover types often emerge after unmixing. In this study, two Landsat 8 Operational Land Imager (OLI) images-acquired on 20 September 2019 (Path/Row: 121/44) and 14...
Multi-scale data have had a wide-ranging level of performance in the area urban change monitoring. Herein we investigate correlation between impervious surface fraction (ISF) and Defense Meteorological Satellite Program/Operational Linescan System (DMSP/OLS) nighttime stable light (NTL) with respect to expansion main districts Guangzhou. Landsat 5 Thematic Mapper 8 Operational Land Image (OLI) from 1988 2015 were used extract ISF using linear spectral mixture analysis model normal difference...
For the large-area snow depth (SD) data sets with high spatial resolution in Altay region of Northern Xinjiang, China, we present a deterministic ensemble Kalman filter (DEnKF)-albedo assimilation scheme that considers common land model (CoLM) subgrid heterogeneity. In albedo DEnKF-albedo, assimilated albedos over each tile are estimated MCD43C1 bidirectional reflectance distribution function (BRDF) parameters product and CoLM calculated solar zenith angle. The BRDF hypothesized to be...
In technologically underdeveloped areas, water pollution threatens the living environment of local residents, so remote sensing monitoring features around reservoirs is necessary. Fully convolutional networks (FCNs) offer great potential for extracting high-resolution due to their unlimited input image size and higher accuracy compared neural networks. Therefore, a proposal classify WorldView-2 images implemented with sixty-eight thousand iterations fine-tuning fully trained combined...