Filling Landsat ETM+ SLC-off gaps using a segmentation model approach

Photogrammetric Engineering and Remote Sensing



The purpose of this article is to present a methodology for filling Landsat Scan Line Corrector (SLC)-off gaps with same-scene spectral data guided by a segmentation model. Failure of the SLC on the Landsat 7 Enhanced Thematic Mapper Plus (ETM+) instrument resulted in a loss of approximately 25 percent of the spectral data. The missing data span across most of the image with scan gaps varying in size from two pixels near the center of the image to 14 pixels along the east and west edges. Even with the scan gaps, the radiometric and geometric qualities of the remaining portions of the image still meet design specifications and therefore contain useful information (see http:// for additional information). The U.S. Geological Survey EROS Data Center (EDC) is evaluating several techniques to fill the gaps in SLC-off data to enhance the usability of the imagery (Howard and Lacasse 2004) (PE&RS, August 2004). The method presented here uses a segmentation model approach that allows for same-scene spectral data to be used to fill the gaps. The segment model is generated from a complete satellite image with no missing spectral data (e.g., Landsat 5, Landsat 7 SLCon, SPOT). The model is overlaid on the Landsat SLC-off image, and the missing data within the gaps are then estimated using SLC-off spectral data that intersect the segment boundary. A major advantage of this approach is that the gaps are filled using spectral data derived from the same SLC-off satellite image.

Publication type Article
Publication Subtype Journal Article
Title Filling Landsat ETM+ SLC-off gaps using a segmentation model approach
Series title Photogrammetric Engineering and Remote Sensing
Volume 70
Issue 10
Year Published 2004
Language English
Publisher ASPRS
Contributing office(s) Earth Resources Observation and Science (EROS) Center
Description 3 p.
First page 1109
Last page 1111
Online Only (Y/N) N
Additional Online Files (Y/N) N
Google Analytic Metrics Metrics page
Additional publication details