卫星遥感图像并行几何校正算法研究

Parallel Algorithm of Geometrical Correction for Satellite Images

几何校正是遥感图像处理过程中的重要环节 ,具有计算量大、耗时长的特点 ,导致遥感图像处理的效率低下 .该文提出一种分布存储环境下的并行几何校正算法 ,每个处理器通过计算本地输入子图像在目标图像中的范围 ,确定其需要进行重采样计算的区域 ,使计算过程中所需的数据均为本地数据 ,很好地解决了数据局部性问题 .文章利用首尾相连的闭线段近似表示理想的输出图像块边界这一思想 ,详细讨论了局部输出区域的计算方法 ,并采用一种新的存储结构用于保存校正后的输出图像块信息 .在机群系统上对算法进行实现 ,结果表明该算法具有良好的并行性能 .

Geometrical correction is an important and computation-intensive task in the processing of remote sensing images. In order to improve its efficiency, this paper provides a parallel geometrical correction algorithm based on distributed memory systems. In the algorithm, each processor calculates the corresponding area in the target image for the local sub input image, and do resampling for this area. This makes all of data needed be in local memory and no communication happens during parallel computing. Closed line segments connected end to end with each other are used to represent the ideal edge of each sub output image approximately when calculating local output area, and a data structure is put forward to save irregular sub output images. By implementing the algorithm on a cluster system, the results show that, this parallel algorithm improves the efficiency of geometrical correction greatly.