allowFullScreen='true' src='http://d.yimg.com/m/up/ypp/default/player.swf' flashvars='vid=21484216&autoPlay=0'>
Keith Wiley, University of Washington, talks about parallel distributed image stacking and mosaicing with Hadoop, and reports on his experience implementing a scalable image-processing pipeline for the SDSS database using Hadoop. This multi-Terabyte imaging dataset provides a good testbed for algorithm development since its scope and structure approximate future surveys. His pipeline performs two primary functions: stacking and mosaicing, in which multiple partially overlapping images are registered, integrated and stitched into a single overarching image. He discusses two implementations, with the latter prepending the Hadoop job with a SQL-based metadata query, thus eliminating the same files from consideration before running the MapReduce job.