Image restoration techniques as applied to Landsat MSS and TM data

Pecora XI Symposium



  • The Publications Warehouse does not have links to digital versions of this publication at this time
  • Download citation as: RIS | Dublin Core


Two factors are primarily responsible for the loss of image sharpness in processing digital Landsat images. The first factor is inherent in the data because the sensor's optics and electronics, along with other sensor elements, blur and smear the data. Digital image restoration can be used to reduce this degradation. The second factor, which further degrades by blurring or aliasing, is the resampling performed during geometric correction.

An image restoration procedure, when used in place of typical resampled techniques, reduces sensor degradation without introducing the artifacts associated with resampling.

The EROS Data Center (EDC) has implemented the restoration proceed for Landsat multispectral scanner (MSS) and thematic mapper (TM) data. This capability, developed at the University of Arizona by Dr. Robert Schowengerdt and Lynette Wood, combines restoration and resampling in a single step to produce geometrically corrected MSS and TM imagery. As with resampling, restoration demands a tradeoff be made between aliasing, which occurs when attempting to extract maximum sharpness from an image, and blurring, which reduces the aliasing problem but sacrifices image sharpness. The restoration procedure used at EDC minimizes these artifacts by being adaptive, tailoring the tradeoff to be optimal for individual images.

Additional publication details

Publication type:
Publication Subtype:
Journal Article
Image restoration techniques as applied to Landsat MSS and TM data
Series title:
Pecora XI Symposium
Year Published:
American Society of Photogrammetry
Publisher location:
Falls Church, VA
1 p.
Larger Work Type:
Larger Work Subtype:
Journal Article
Larger Work Title:
Pecora XI Symposium
First page:
Last page: