An effective gap interpolation and filtering (GIF) method for landsat 7 ETM+ SLC-off Data

Abstract

The Landsat 7 Enhanced Thematic Mapper 23 Plus (ETM+) scan line 24 corrector (SLC) failed on May 31, 2003, causing SLC turned off. Many gap-filled 25 products were developed and deployed to combat this situation. The majority of these 26 used a primary image taken by the SLC when functioning properly in an attempt to correct 27 an SLC-off image. However, by using a primary image, temporal atmospheric elements 28 could not be reliably reflected, and therefore the corrected image was not viable for use 29 by monitoring systems. To bypass this limitation, this study has developed the GIF 30 method that relies merely on the use of one-dimensional interpolation filtering to 31 conveniently recover pixels within a single image at a high level of accuracy without 32 borrowing from images acquired at a different time or by another sensor. To analyze the 33 accuracy of the GIF method, it was compared to two other methods-GLHM and 34 LLHM-both developed by NASA and USGS. The accuracy of the GIF method came 35 out superior in its images of land, sea, and clouds. In particular, its images of sea and 36 clouds returned RSME values of close to or less than 1. We expect the GIF method 37 developed through this research to be of invaluable support to monitoring systems that 38 depend heavily on Landsat imagery.