Share Email Print

Proceedings Paper

Resampling satellite scenes
Author(s): John Stokes
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Procedures used for resampling digital images usually assume the underlying function to be band limited, have continuous third derivatives or the like. When objects are present in an image, these requirements are strongly violated. Object boundaries must then be considered. A resampling procedure is designed which starts with a segmentation in order to identify regions which are continuous and smooth. The segment interiors are modelled in a least squares adjustment determining the parameters of a low order polynomial and a residual noise component. Segment boundaries are modelled so as to conform with our concepts of objects. Interpolation is then performed separately in each segment, only using data belonging to the segment. Properties of satellite data are discussed and used as assumptions for deriving an appropriate interpolator. A set of procedures fulfilling these assumptions is described, the simplest procedure being cubic convolution. The tests performed show that the method can handle strong grey level discontinuities. The tests on real scenes are however inconclusive, as discontinuities are smoothed during the production phase.

Paper Details

Date Published: 17 August 1994
PDF: 8 pages
Proc. SPIE 2357, ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision, (17 August 1994); doi: 10.1117/12.182867
Show Author Affiliations
John Stokes, Royal Institute of Technology (Sweden)

Published in SPIE Proceedings Vol. 2357:
ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision
Heinrich Ebner; Christian Heipke; Konrad Eder, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?