A coastal algorithm for fully automatic geometric correction of Advanced Very High Resolution Radiometer (AVHRR) images is presented. Inputs are the AVHRR image and updated ephemeris data and outputs are the georeference image and a cloud image mask. Its principal advantage and novelty is that it requires only manual control in the first stage of the process. Particularly, the detection of Ground Control Points (GCPs), usually rather time consuming, is performed with this method in an automatic way. The procedure only requires the previous existence of a coastal reference-windows database. To find the exact location of the GCPs, the routine searches the best match of these referencewindows with the image. The complete automation of the process makes the routine very fast, then allowing its operative application on a large volume of images. The process provides accuracies to within 1-1.5 AVHRR pixels.
[1]
R. J. H. Brush,et al.
The navigation of AVHRR imagery
,
1988
.
[2]
Diem Ho,et al.
NOAA AVHRR image referencing
,
1986
.
[3]
M. Derrien,et al.
Automatic cloud detection applied to NOAA-11 /AVHRR imagery
,
1993
.
[4]
M. Cantón,et al.
The spatial and temporal evolution of thermal structures in the Alborán Sea Mediterranean basin
,
1998
.
[5]
M. Cantón,et al.
Sea surface temperature variability in Alboran sea from satellite data
,
1998
.
[6]
José F. Moreno,et al.
An optimum interpolation method applied to the resampling of NOAA AVHRR data
,
1994,
IEEE Trans. Geosci. Remote. Sens..
[7]
Jörg Bendix,et al.
An improved algorithm for NOAA-AVHRR image referencing
,
1992
.