李鹏飞, 孙开敏, 李德仁, 王玮. 无人机影像应急并行处理负载均衡方法[J]. 武汉大学学报 ( 信息科学版), 2018, 43(2): 268-274. DOI: 10.13203/j.whugis20170022
引用本文: 李鹏飞, 孙开敏, 李德仁, 王玮. 无人机影像应急并行处理负载均衡方法[J]. 武汉大学学报 ( 信息科学版), 2018, 43(2): 268-274. DOI: 10.13203/j.whugis20170022
LI Pengfei, SUN Kaimin, LI Deren, WANG Wei. A Load Balancing Strategy for Urgent Parallel Processing of UAV Imagery[J]. Geomatics and Information Science of Wuhan University, 2018, 43(2): 268-274. DOI: 10.13203/j.whugis20170022
Citation: LI Pengfei, SUN Kaimin, LI Deren, WANG Wei. A Load Balancing Strategy for Urgent Parallel Processing of UAV Imagery[J]. Geomatics and Information Science of Wuhan University, 2018, 43(2): 268-274. DOI: 10.13203/j.whugis20170022

无人机影像应急并行处理负载均衡方法

A Load Balancing Strategy for Urgent Parallel Processing of UAV Imagery

  • 摘要: 基于应急条件下的计算机和网络资源,以实现无人机影像应急并行处理为目的,提出了一种简洁的负载均衡实现方法。采用了以进程为调度单元的主动式计算资源统计策略,辅以任务队列二级动态分配方法,实现了负载均衡。使用本方法在应急环境中建立简易的并行化集群处理系统,可以灵活、高效、鲁棒地利用可用的计算资源,大幅缩短无人机影像应急处理时间。为验证本文负载均衡方法的有效性,利用无人机机载定位定向系统记录的定位定向参数,结合已知数字高程数据,完成了无人机影像的初步正射纠正和拼图处理。实验结果表明,该方法实现简单、部署方便、扩展灵活、任务分配均衡,处理效率随着计算单元的增加而提高,可大幅提高应急条件下无人机应急影像图制作效率。

     

    Abstract: An implementation of a simplified load balancing scheme was investigated to provide a paralleling way of processing Unmanned Aerial Vehicle (UAV) imagery based on common computing and network resources for urgent response. A load balancing strategy, which actively monitors computing resources at the process-level and dynamically manages two-stage mission queues, was adopted. Building a lite parallel processing system under urgent circumstances using this method will provide a flexible, efficient, and robust way to accelerate UAV imagery production markedly, based on available computing resources. Orthorectification and mosaicing of UAV imagery using position and orientation parameters provided by onboard positioning and orientation systems using digital elevation data was carried out to validate the effectiveness of this method. Results show that the proposed solution is easy to implement, convenient to deploy, flexible, and extends and rationalizes mission allocation, leading to an strong correlations between computing nodes and efficiency, thus improving orthophotomap productivity when using UAV imagery under urgent circumstances.

     

/

返回文章
返回