RPC模型影像校正并行算法设计及优化

Design and optimization of parallel algorithm for RPC correction of satellite images

随着遥感技术的发展,卫星遥感影像在环境监测、城市规划、灾害评估、气候变化、农业监测等多个领域的应用越来越广泛。为了应对大规模遥感数据处理带来的计算挑战,提出了基于GPU及多核CPU实现卫星遥感影像RPC几何校正的并行算法,并采用了多流优化、I/O优化、内存分配优化等策略。实验表明,与串行算法相比,GPU并行算法获得了22.9~71.5的计算加速比及2.4~32.4的综合加速比,多核CPU并行算法获得了5.4~5.7的计算加速比及2.1~5.0的综合加速比。

With the development of remote sensing technology,satellite remote sensing images are increasingly used in many fields such as environmental monitoring,urban planning,disaster assessment,climate change,and agricultural monitoring.In order to cope with the computational challenges brought by large-scale remote sensing data processing,this paper proposes a parallel algorithm for RPC geometry correction of satellite remote sensing images based on GPU and multi-core CPU,and adopts strategies such as multi-stream optimization,I/O optimization and memory allocation optimization.Experiments show that compared with the serial algorithm,the GPU parallel algorithm obtains a computational speedup ratio of 22.9~71.5 and a comprehensive speedup ratio of 2.4~32.4,and the multi-core CPU parallel algorithm obtains a computational speedup ratio of 5.4~5.7 and a comprehensive speedup ratio of 2.1~5.0.