Parallel Algorithm Design for Remote Sensing Image Processing in the PC Cluster Environment

摘要

In the PC cluster environment, parallel algorithms can significantly improve the efficiency of remote sensing image processing. The remote sensing dataset is the raster data stored by order of band, therefore, it is feasible to assign executable tasks to some computing nodes by band and complete the processing tasks together through communicating each other amount the computing nodes by MPI interface. In addition, the multi-thread processing algorithm is scheduled based on OpenMP library toward the single computing node with multi-core CPU. The experiment evaluates image radiation performance about the four-band HJ-1A CCD remote sensing image in the PC cluster environment which consists of 4 sets of dual-core computers. Its performance is increased by 6 times to 6.5 times comparing with a single PC. Through validating, the dual level design pattern is available based on integrating MPI and OpenMP to improve the efficiency of remote sensing image processing.