面向剖分面片模板的遥感影像并行处理方法

发布时间：2018-03-24 09:16

本文选题：遥感影像　切入点：地球剖分理论　出处：《计算机应用研究》2016年08期

【摘要】：空间数据特别是遥感影像数据的快速增加和应用需求的扩大,其组织效率和处理速度已经成为制约技术应用的瓶颈,地球剖分理论和高性能计算为上述问题的解决提供了一种可能途径。针对上述问题,在遥感影像剖分面片数据模型的研究基础上,提出了剖分面片模板并行计算模式,设计并实现了一种面向剖分面片模板的遥感影像并行处理方法。该方法基于MPI(message passing interface)与Open MP(open multi-processing)混合并行计算框架,构建算法并行处理模型,形成算法并行化类库,通过调用其内部方法实现计算任务的并行执行。通过一个遥感影像剖分化并行分割处理实例,验证了该方法的有效性。实验结果表明,该方法具有较好的分割效果和加速比,有一定的示范意义,为进一步提高遥感影像应用能力提供了借鉴。
[Abstract]:With the rapid increase of spatial data, especially remote sensing image data, and the expansion of application requirements, the organizational efficiency and processing speed of spatial data, especially remote sensing image data, have become the bottleneck restricting the application of technology. The theory of earth division and high performance calculation provide a possible way to solve the above problems. Based on the research of remote sensing image slicing data model, a parallel computing model of sliced slice template is proposed. A parallel processing method for remote sensing image is designed and implemented, which is based on the hybrid parallel computing framework of MPI(message passing interface and Open MP(open multi-processing. The parallel processing model of the algorithm is constructed and the class library of algorithm parallelization is formed. The parallel execution of computing task is realized by calling its internal method. The effectiveness of this method is verified by an example of partition parallel segmentation of remote sensing image. The experimental results show that the method has better segmentation effect and speedup. It has certain demonstration significance and provides a reference for further improving the ability of remote sensing image application.
【作者单位】：许昌学院信息工程学院;成都理工大学地球物理学院;
【基金】：国家自然科学基金资助项目(U1304403) 河南省科技攻关计划资助项目(132102210398) 河南省基础与前沿技术研究计划资助项目(132300410349)
【分类号】：TP751

【相似文献】