Traditionally,remote sensing image processing is achieved by CPU alone,which is too time-consuming to ensure real-time processing.Recently,the rapid increase in the performance of graphics processing units (GPU) has made it into a compelling platform for image processing.The hybrid CPU-GPU parallel system is presented to accelerate the processing.Meanwhile,in order to fully exploit the potential capacity of the heterogeneous computing platform,a novel sub-aperture approach is proposed.%对于大规模遥感数据处理,单纯依靠CPU实现实时成像处理存在较大挑战.近年来出现的GPU图像处理单元为大规模数据处理提供了一个理想的处理平台.结合GPU和CPU的特点,利用基于CPU-GPU的异构协同处理平台来加速图像处理.同时,为了充分发挥该异质架构的性能,提出了一种改进的基于子孔径架构的成像算法,并在此架构上对超大规模遥感数据进行了成像处理,取得良好的成像效果和加速比.
展开▼