基于复小波变换的遥感图像并行融合算法

随着遥感技术的快速发展，多源遥感图像的快速融合成为很多遥感任务的关键处理步骤。为了加速遥感图像的融合处理过程，本文首先提出了一种新的基于双树复小波变换的并行融合算法（PACWT）。算法中综合运用了数据分布、并行数据处理和负载均衡技术，以克服单机处理在计算能力和存储器空间上的限制；针对基于CWT的图像融合处理的计
计算特点，设计了一种可有效避免计算过程中数据通信的冗余划分方法。然后，从理论上分析了算法在时间和空间两方面的性能。最后，通过实验分析了算法在32-CPU的Cluster系统上的实际性能。结果表明，本文提出的算法具有良好的可扩展性，在数据量较大时可获得良好的加速比和并行效率。

A Parallel Fusion Algorithm for Remote Sensing Images Based on Complex Wavelet Transform

With the rapid development of the remote sensing technology,fast fusion of digital images from disparate sources has become critical to the success of these endeavors. In this paper, to speedup the fusion process, a Parallel Fusion Algorithm based on Dual-Tree Complex Wavelet Transform (PACWT for short) of remote sensing images is presented for the first time. To overcome the limitations on the memory space as well as the computing capability of a single processor, data distribution, data-parallel processing and load balancing techniques are integrated into PACWT. To avoid the inherent communication overhead of the complex-wavelet-based fusion method, a special design called redundant partitioning is used, which is inspired by the characteristics of complex wavelet transform. Finally, PACWT is evaluated in theory and tested on a 32-CPU cluster of workstations. The experimental results show that our algorithm has good parallel performance and scalability.