Research On Parallel Slicing Method For Five-Layer Fifteen-Level Remote Sensing Tiles And The Design Of GCF Storage

The five-layer fifteen-level remote sensing data organization model provides a new standard for remote sensing image data storage and service.This model standardizes remote sensing data,establishes a unified segmentation model and hierarchical image resolution,designs a standardized data structure,retains the basic information of the original remote sensing image data,shields the differences between multi-source remote sensing data,and finally realizes the integrated organization and management of remote sensing image data,supports the comprehensive processing and collaborative application of multi-source and multi-resolution remote sensing data,and provides standardized remote sensing data organization foundation and technical support.As the data foundation for various related applications and comprehensive processing,there is little research on the tiling and storage format of five-layer fifteenlevel remote sensing tiles.Based on the principles and characteristics of five-layer fifteen-level remote sensing data organization model,this paper proposes a five-layer fifteen-level remote sensing tile parallel slicing method suitable for different scenarios,and designs a five-layer fifteen-level remote sensing tile storage The format verifies the effectiveness of the method through the actual application of the remote sensing big data three-dimensional visualization platform.The specific research content is as follows:(1)Based on the five-layer fifteen-level segmentation model and hierarchical standards,and combined with the principles and characteristics of parallel computing,a queuing polling method and a hierarchical classification and division method are proposed.Multi-level slicing experiments show that this method can improve the efficiency of single-machine slicing of a single image.In order to improve the efficiency of multi-level slicing of images with large amounts of data,combined with Kafka to achieve multi-level distributed slicing of multiple images,the slicing efficiency can be increased by about three times.(2)A GCF storage format suitable for five-layer fifteen-level remote sensing tiles is designed,and a hierarchical and block organization model is proposed.The model follows the five-layer fifteen-level organization standard,and uniquely encodes the remote sensing tile data block,and realizes the hierarchical and block organization of the five-layer fifteen-level remote sensing tile data through the secondary index,and the tile data is formed according to the level Big data set file.(3)According to the GCF data storage and reading and writing design,a GCFbased target area tile acquisition method is proposed to achieve rapid acquisition of a set of target tile data blocks at a specific level and specific spatial range.On this basis,the construction of a three-dimensional visualization platform for remote sensing big data was realized,and the functions of multi-time comparison,time series and rolling shutter visualization of five-layer fifteen-level remote sensing tile data were realized,and the effectiveness of the GCF storage was verified.