Pinned Repositories
cuESTARFM
cuESTARFM is a GPU-enabled enhanced Spatial and Temporal Adaptive Reflectance Fusion Model (ESTARFM)
cuFSDAF
cuFSDAF is an enhanced FSDAF algorithm parallelized using GPUs. In cuFSDAF, the TPS interpolator is replaced by a modified Inverse Distance Weighted (IDW) interpolator. Besides, computationally intensive procedures are parallelized using the Compute Unified Device Architecture (CUDA), a parallel computing framework for GPUs. Moreover, an adaptive domain-decomposition method is developed to adjust the size of sub-domains according to hardware properties adaptively and ensure the accuracy at the edges of sub-domains.
cuSTARFM
cuSTARFM is a GPU-enabled Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM)
cuSTNLFFM
cuSTNLFFM is a GPU-enabled Spatial and Temporal Non-Local Filter-based Fusion Model (STNLFFM)
cuSTSG
cuSTSG is a GPU-enabled spatial-temporal Savitzky-Golay (STSG) program based on the Compute Unified Device Architecture (CUDA). Firstly, the cosine similarities between the annual NDVI time series are used to identify and exclude the NDVI values with inaccurate quality flags from the NDVI seasonal growth trajectory. Secondly, the computational performance is improved by reducing redundant computations, and parallelizing the computationally intensive procedures using CUDA on GPUs.
Mixed_Cell_Cellullar_Automata
The Mixed-Cell Cellullar Automata (MCCA) provides a new approach to enable more dynamic mixed landuse modeling to move away from the analysis of static patterns. One of the biggest advantages of mixed-cell CA models is the capability of simulating the quantitative and continuous changes of multiple landuse components inside cells.
MSTGAN-airquality-prediction
MSTGAN is an innovative method designed for multi-station urban air quality prediction, which fully considers the individual, global, and local multi-scale information of air quality spatiotemporal sequences. It incorporates an attention-based dynamic graph modeling approach to capture global spatiotemporal dependencies.
Open-Space-Cellular_Automata
A spatio-temporal approach based on Cellular Automata (CA) for simulating the spatial dynamics of open spaces (include urban green spaces, parks, squares, trails, courtyards, and other natural spaces), by considering a set of spatial data that represents the infrastructural and socio-economic factors, namely the OS-CA (Open Space Cellular Automaton) model. The dynamic sub-model for OS is used to generate scenarios with different parameters (e.g. mean construction delays and mean area) for exploring the effects of planning policies on the future distribution of open space. The OS-CA considers the interactions and inter-attraction between open space and urban land in the simulation process. The proposed model can accurately predict the emergence of some open spaces.
Patch-generating_Land_Use_Simulation_Model
The PLUS model integrates a rule mining framework based on Land Expansion Analysis Strategy (LEAS) and a CA model based on multi-type Random Patch Seeds (CARS), which was used to understand the drivers of land expansion and project landscape dynamics.
pRPL
parallel Raster Processing Library (pRPL) is a MPI-enabled C++ programming library that provides easy-to-use interfaces to parallelize raster/image processing algorithms
High-performance Spatial Computational Intelligence Lab @ China University of Geosciences (Wuhan)'s Repositories
HPSCIL/Patch-generating_Land_Use_Simulation_Model
The PLUS model integrates a rule mining framework based on Land Expansion Analysis Strategy (LEAS) and a CA model based on multi-type Random Patch Seeds (CARS), which was used to understand the drivers of land expansion and project landscape dynamics.
HPSCIL/cuSTARFM
cuSTARFM is a GPU-enabled Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM)
HPSCIL/cuESTARFM
cuESTARFM is a GPU-enabled enhanced Spatial and Temporal Adaptive Reflectance Fusion Model (ESTARFM)
HPSCIL/Mixed_Cell_Cellullar_Automata
The Mixed-Cell Cellullar Automata (MCCA) provides a new approach to enable more dynamic mixed landuse modeling to move away from the analysis of static patterns. One of the biggest advantages of mixed-cell CA models is the capability of simulating the quantitative and continuous changes of multiple landuse components inside cells.
HPSCIL/cuFSDAF
cuFSDAF is an enhanced FSDAF algorithm parallelized using GPUs. In cuFSDAF, the TPS interpolator is replaced by a modified Inverse Distance Weighted (IDW) interpolator. Besides, computationally intensive procedures are parallelized using the Compute Unified Device Architecture (CUDA), a parallel computing framework for GPUs. Moreover, an adaptive domain-decomposition method is developed to adjust the size of sub-domains according to hardware properties adaptively and ensure the accuracy at the edges of sub-domains.
HPSCIL/Open-Space-Cellular_Automata
A spatio-temporal approach based on Cellular Automata (CA) for simulating the spatial dynamics of open spaces (include urban green spaces, parks, squares, trails, courtyards, and other natural spaces), by considering a set of spatial data that represents the infrastructural and socio-economic factors, namely the OS-CA (Open Space Cellular Automaton) model. The dynamic sub-model for OS is used to generate scenarios with different parameters (e.g. mean construction delays and mean area) for exploring the effects of planning policies on the future distribution of open space. The OS-CA considers the interactions and inter-attraction between open space and urban land in the simulation process. The proposed model can accurately predict the emergence of some open spaces.
HPSCIL/cuSTNLFFM
cuSTNLFFM is a GPU-enabled Spatial and Temporal Non-Local Filter-based Fusion Model (STNLFFM)
HPSCIL/pRPL
parallel Raster Processing Library (pRPL) is a MPI-enabled C++ programming library that provides easy-to-use interfaces to parallelize raster/image processing algorithms
HPSCIL/cuSTSG
cuSTSG is a GPU-enabled spatial-temporal Savitzky-Golay (STSG) program based on the Compute Unified Device Architecture (CUDA). Firstly, the cosine similarities between the annual NDVI time series are used to identify and exclude the NDVI values with inaccurate quality flags from the NDVI seasonal growth trajectory. Secondly, the computational performance is improved by reducing redundant computations, and parallelizing the computationally intensive procedures using CUDA on GPUs.
HPSCIL/MSTGAN-airquality-prediction
MSTGAN is an innovative method designed for multi-station urban air quality prediction, which fully considers the individual, global, and local multi-scale information of air quality spatiotemporal sequences. It incorporates an attention-based dynamic graph modeling approach to capture global spatiotemporal dependencies.
HPSCIL/mcRPL
HPSCIL/CSD-RkNN
CSD-RkNN: Conic Section Discriminances for Large Scale Reverse k Nearest Neighbors Queries
HPSCIL/SCMA-MCAE
A Spatially Constrained Multi-Autoencoder Approach for Multivariate Geochemical Anomaly Recognition and a MCAE approach for multivariate geochemical anomaly recognition
HPSCIL/mcPLUS
HPSCIL/mcMCCA
HPSCIL/rupMC
rupMC is a parallel Marching Cubes (MC) program based on the ray-unit. Firstly, ray-units are used in rupMC as the basic voxel to determine how the surface intersects. Secondly, rupMC uses multiple computing processes and threads on a CPU/GPU heterogeneous architecture to process the points concurrently.
HPSCIL/ASTS
Adaptive-Spatiotemporal-Sampling
HPSCIL/CHN6-CUG-Roads-Dataset
HPSCIL/intCBDS
HPSCIL/SSDGL
SSDGL: A Spectral-Spatial-Dependent Global Learning Framework for Insufficient and Imbalanced Hyperspectral Image Classification (TCYB2021) https://ieeexplore.ieee.org/document/9440852