Abstract
Grid computing seeks to aggregate computing resources within an enterprise and leverage on resources you don?t own for compute-intensive applications. Geo-rectification is a process for correcting spatial location and orientation of a satellite image. This paper focuses on the parallelization of the compute-intensive satellite image geo-rectification problem on a cluster grid. We discuss our approach to data and task partitioning, visualization technique and the archival of data. The computational tasks include wrapping satellite positional data to compensate the earth curvature, and consist of several steps such as image re-sampling, resolution conversion and image matching. Experimental results obtained using commodity PCs are discussed.