Building large scale mosiacs from Landsat data

  • Authors:

  • Lucian Plesea;Joseph Jacob

  • Affiliations:

  • Jet Propulsion Laboratory, 4800 Oak Grove Drive, MS 126-234, Pasadena, CA;Jet Propulsion Laboratory, 4800 Oak Grove Drive, MS 126-234, Pasadena, CA

  • Venue:
  • Proceedings of the 8th ACM international symposium on Advances in geographic information systems
  • Year:
  • 2000

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper describes the process and tools used for creating a full resolution seamless Landsat mosaic of the continental U.S. at 30 meters per pixel. The 6-band, 150GB image was built from 428 individual Landsat scenes of the Multi-Resolution Land Characteristics (MRLC) dataset. Custom designed software based on SGI ImageVision framework was developed on JPL supercomputers to accomplish this task as a single step operation, with minimal operator input. Novel techniques were used to deal with data storage and access issues. Various color matching algorithms can be used to generate a parameter set, which in turn controls all the details of building the mosaic. Some of the tested algorithms and their results will also be discussed.