In the design of a graphic processing unit(GPU),the processing speed of triangle rasterization is an important factor that determines the performance of the GPU.An architecture of a multi-tile parallel-scan rasterizat...In the design of a graphic processing unit(GPU),the processing speed of triangle rasterization is an important factor that determines the performance of the GPU.An architecture of a multi-tile parallel-scan rasterization accelerator was proposed in this paper.The accelerator uses a bounding box algorithm to improve scanning efficiency.It rasterizes multiple tiles in parallel and scans multiple lines at the same time within each tile.This highly parallel approach drastically improves the performance of rasterization.Using the 65 nm process standard cell library of Semiconductor Manufacturing International Corporation(SMIC),the accelerator can be synthesized to a maximum clock frequency of 220 MHz.An implementation on the Genesys2 field programmable gate array(FPGA)board fully verifies the functionality of the accelerator.The implementation shows a significant improvement in rendering speed and efficiency and proves its suitability for high-performance rasterization.展开更多
Snow water equivalent(SWE)is an important factor reflecting the variability of snow.It is important to estimate SWE based on remote sensing data while taking spatial autocorrelation into account.Based on the segmentat...Snow water equivalent(SWE)is an important factor reflecting the variability of snow.It is important to estimate SWE based on remote sensing data while taking spatial autocorrelation into account.Based on the segmentation method,the relationship between SWE and environmental factors in the central part of the Tibetan Plateau was explored using the eigenvector spatial filtering(ESF)regression model,and the influence of different factors on the SWE was explored.Three sizes of 16×16,24×24 and 32×32 were selected to segment raster datasets into blocks.The eigenvectors of the spatial adjacency matrix of the segmented size were selected to be added into the model as spatial factors,and the ESF regression model was constructed for each block in parallel.Results show that precipitation has a great influence on SWE,while surface temperature and NDVI have little influence.Air temperature,elevation and surface temperature have completely different effects in different areas.Compared with the ordinary least square(OLS)linear regression model,geographically weighted regression(GWR)model,spatial lag model(SLM)and spatial error model(SEM),ESF model can eliminate spatial autocorrelation with the highest accuracy.As the segmentation size increases,the complexity of ESF model increases,but the accuracy is improved.展开更多
基金the Scientific Research Program Funded by Shaanxi Provincial Education Department(20JY058)。
文摘In the design of a graphic processing unit(GPU),the processing speed of triangle rasterization is an important factor that determines the performance of the GPU.An architecture of a multi-tile parallel-scan rasterization accelerator was proposed in this paper.The accelerator uses a bounding box algorithm to improve scanning efficiency.It rasterizes multiple tiles in parallel and scans multiple lines at the same time within each tile.This highly parallel approach drastically improves the performance of rasterization.Using the 65 nm process standard cell library of Semiconductor Manufacturing International Corporation(SMIC),the accelerator can be synthesized to a maximum clock frequency of 220 MHz.An implementation on the Genesys2 field programmable gate array(FPGA)board fully verifies the functionality of the accelerator.The implementation shows a significant improvement in rendering speed and efficiency and proves its suitability for high-performance rasterization.
基金funded by the National Key S&T Special Projects of China(grant number:2018YFB0505302)the National Nature Science Foundation of China(grant number:41671380)。
文摘Snow water equivalent(SWE)is an important factor reflecting the variability of snow.It is important to estimate SWE based on remote sensing data while taking spatial autocorrelation into account.Based on the segmentation method,the relationship between SWE and environmental factors in the central part of the Tibetan Plateau was explored using the eigenvector spatial filtering(ESF)regression model,and the influence of different factors on the SWE was explored.Three sizes of 16×16,24×24 and 32×32 were selected to segment raster datasets into blocks.The eigenvectors of the spatial adjacency matrix of the segmented size were selected to be added into the model as spatial factors,and the ESF regression model was constructed for each block in parallel.Results show that precipitation has a great influence on SWE,while surface temperature and NDVI have little influence.Air temperature,elevation and surface temperature have completely different effects in different areas.Compared with the ordinary least square(OLS)linear regression model,geographically weighted regression(GWR)model,spatial lag model(SLM)and spatial error model(SEM),ESF model can eliminate spatial autocorrelation with the highest accuracy.As the segmentation size increases,the complexity of ESF model increases,but the accuracy is improved.
