×

You are using an outdated browser that does not fully support the intranda viewer.
As a result, some pages may not be displayed correctly.

We recommend you use one of the following browsers:

Full text

Title
Proceedings of the Symposium on Global and Environmental Monitoring

Correction for Off-Nadir Viewing and Topographic Effects
Sun j i Mural, Toshiaki Hashimoto
Institute of Industrial Science, University of Tokyo
7-22 Roppongi, Minatoku, Tokyo
Japan
Abstract
Off-nadir viewing causes some geometric distortions in imagery
like a tangential distortion, etc. Furthermore, foreshortening or
layover may happen in case of the off-nadir viewing of an imaging
radar.
This paper describes the correction for such kinds of distor
tions owing to off-nadir viewing and topogragraphic effects. And
is also introduced the generation of DTM from SPOT stereo pairs.
1. Introduction
Here are summarized some kinds of results of studies made by
the authors.
The algorithm of geometric correction based on the photogram
me trie principles has been developed. This can be adopted to both
high resolution imagery like LANUSAT MSS |1I and low one like MOS-1
VTIR l2 J .
SAR imagery is distorted in mountainous areas because of fore
shortening or layover. The distortions can be eliminated with DTMs
13 J .
Since the launch of SPOT, the interest of three dimensional
measurement; from space has been increasing. The algorithm developed
by the authors for DTM generation from SPOT stereo pairs is also
introduced [4] [5].
2. Geometric Correction of LANDSAT MSS Imagery
The methodologies of geometric correction of satellite imagery
are classified into three; 1) system correction with satellite data
only, 2) black box correction with GCPs ( ground control points )
only and 3) combined method of system correction and black box one.
It: has been pointed out that the third method is the best in
accuracy and computing time. However, this method is a little bit
complicated since it is composed of two steps process. The simul
taneous process has been developed by treating the position and
attitude of sensor as the unknown variables, which are determined
from colinearity equations in photogramme t; ry.
The position and attitude of sensor can be regarded as exteri
or orientation parameters in photogramme try. These parameters are a
function of time or line number since the scanning is performed
every short time. Although the scanning geometry of MSS is not a
central projection, it can be treated like central projection when
800